coilyco-gaming/sirens-echo
Watch
3
0
Code Issues 153 Pull requests Projects Releases Packages 2 Wiki Activity Actions

when the essence of a question is primarily the knowledge contents of a wiki page, echo should link to that wiki page #222

New issue
Open
opened 2026-08-13 01:42:15 +00:00 by coilysiren · 9 comments
coilysiren commented 2026-08-13 01:42:15 +00:00
Owner
Copy link

example prompt

@sirens-echo while you’re busy not being a cow, can you tell me what algebra is?

Algebra is a branch of mathematics that uses symbols and letters to represent numbers and quantities in equations and formulas. It allows for the general formulation of arithmetic operations and relationships between variables.

Key components include:
Variables (letters like x, y, z representing unknown values)
Constants (fixed numerical values)
Operators (+, -, ×, ÷)
Equations (statements that two expressions are equal)

Algebra enables problem-solving by establishing mathematical relationships and finding unknown values through systematic manipulation of equations. It forms the foundation for more advanced mathematics including calculus, linear algebra, and abstract algebra.

reply should include a link to https://en.wikipedia.org/wiki/Algebra

example prompt --- @sirens-echo while you’re busy not being a cow, can you tell me what algebra is? --- Algebra is a branch of mathematics that uses symbols and letters to represent numbers and quantities in equations and formulas. It allows for the general formulation of arithmetic operations and relationships between variables. Key components include: Variables (letters like x, y, z representing unknown values) Constants (fixed numerical values) Operators (+, -, ×, ÷) Equations (statements that two expressions are equal) Algebra enables problem-solving by establishing mathematical relationships and finding unknown values through systematic manipulation of equations. It forms the foundation for more advanced mathematics including calculus, linear algebra, and abstract algebra. --- reply should include a link to https://en.wikipedia.org/wiki/Algebra
coilyco-ops commented 2026-08-13 03:25:53 +00:00
Member
Copy link

CLAIM — Lucia (AI) at 2026-08-13T03:32Z, 20 minute hold. Folded into the outbound-link cluster anchored at #213 — see that thread for the plan and the scope line.

**CLAIM — Lucia (AI)** at 2026-08-13T03:32Z, 20 minute hold. Folded into the outbound-link cluster anchored at https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/213 — see that thread for the plan and the scope line.
coilyco-ops commented 2026-08-13 03:46:03 +00:00
Member
Copy link

Design decision — general link-out rule

Recorded by Delphi (design seat). Kai's decision, 2026-08-12.

Decided. When a question is essentially "explain this encyclopedia topic," Echo gives a brief direct answer and links an authoritative source, rather than reproducing the page in prose. The algebra reply quoted above is the anti-pattern.

Kai explicitly kept these questions in scope — she rejected making general-knowledge explainers a classifier denylist category (#227). Echo still answers. It just stops performing as the encyclopedia and hands off to the source.

Implementation rides with the curated-URL work in #213, which is the anchor for the whole linking regime — the Eco curated list and this general rule are two mechanisms under one behavior. Open questions about source selection are recorded there.

Eval case: the verbatim algebra prompt above. Pass = brief answer plus authoritative link. Fail = the current wall of prose.

## Design decision — general link-out rule Recorded by Delphi (design seat). Kai's decision, 2026-08-12. **Decided.** When a question is essentially "explain this encyclopedia topic," Echo gives a **brief** direct answer and links an authoritative source, rather than reproducing the page in prose. The algebra reply quoted above is the anti-pattern. Kai explicitly kept these questions **in scope** — she rejected making general-knowledge explainers a classifier denylist category (https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/227). Echo still answers. It just stops performing as the encyclopedia and hands off to the source. Implementation rides with the curated-URL work in https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/213, which is the anchor for the whole linking regime — the Eco curated list and this general rule are two mechanisms under one behavior. Open questions about source selection are recorded there. **Eval case:** the verbatim algebra prompt above. Pass = brief answer plus authoritative link. Fail = the current wall of prose.
coilysiren commented 2026-08-13 03:55:12 +00:00
Author
Owner
Copy link

reopening to double check delphi's clarification was addressed

reopening to double check delphi's clarification was addressed
coilyco-ops commented 2026-08-13 07:46:24 +00:00
Member
Copy link

Audit against Delphi's clarification, which is what you reopened this for — Lucia (AI). a4e896c. Four hours late, and it was my claim, so the delay is mine.

Point 1, brief answer plus authoritative link rather than reproducing the page. Holds. links-community.md carries a General reference section that covers exactly this shape:

For a question whose substance is ordinary encyclopedic knowledge rather than Eco or this community, the English Wikipedia article for the plain subject of the question is an approved link.

Answer briefly and let the link carry the depth. Two or three sentences, then the link. A full explainer with a link attached is the failure this rule prevents, not a thorough version of it.

That last sentence is your anti-pattern named directly. One correction to my own record: I went looking for this in links-eco.md and it is not there. My recall of my own file was wrong, not the file.

Point 2, general-knowledge stays in scope rather than becoming a denylist category. Holds. agent/content-classes.yaml lists it as allowed, and boundaries.md does not deny it. Echo still answers.

Point 3, the eval case. Did not exist. It does now.

In the rate pack rather than the gate, because "how often is this followed" is a rate question, and a gate would turn one bad run into a failed deployment. Both halves are measured, using your line "reply should include a link to https://en.wikipedia.org/wiki/Algebra":

WALL    89 words, no link   -> fails length AND required pattern
GOOD    35 words, link      -> passes both
TERSE   16 words, link      -> passes both
NOLINK  15 words, no link   -> fails required pattern only

The reply you quoted is 89 words and a compliant one is 16 to 35, so the 70 word ceiling sits between them with margin in both directions rather than against either edge.

The NOLINK row is the one your line actually turns on. A brief, correct, unlinked answer fails. That is a stronger reading than brevity alone and it is what you asked for.

One thing I want on the record, because I have argued against required patterns all night. They are the dangerous kind of check: no single expression accepts every correct phrasing. This one is safe for a specific reason — the doctrine builds this address rather than copying it, as the plain article form from the subject the member named, so for this prompt exactly one URL is correct. I would not have written a check requiring "some link", and I would push back on one.

Not closing. The case has never been run, since nothing has reached a live model from here (#249). The rule is written, the measurement exists, and the number does not.

**Audit against Delphi's clarification, which is what you reopened this for — Lucia (AI).** `a4e896c`. Four hours late, and it was my claim, so the delay is mine. **Point 1, brief answer plus authoritative link rather than reproducing the page. Holds.** `links-community.md` carries a General reference section that covers exactly this shape: > For a question whose substance is ordinary encyclopedic knowledge rather than Eco or this community, the English Wikipedia article for the plain subject of the question is an approved link. > > **Answer briefly and let the link carry the depth.** Two or three sentences, then the link. A full explainer with a link attached is the failure this rule prevents, not a thorough version of it. That last sentence is your anti-pattern named directly. **One correction to my own record:** I went looking for this in `links-eco.md` and it is not there. My recall of my own file was wrong, not the file. **Point 2, general-knowledge stays in scope rather than becoming a denylist category. Holds.** `agent/content-classes.yaml` lists it as allowed, and `boundaries.md` does not deny it. Echo still answers. **Point 3, the eval case. Did not exist. It does now.** In the rate pack rather than the gate, because "how often is this followed" is a rate question, and a gate would turn one bad run into a failed deployment. Both halves are measured, using your line *"reply should include a link to https://en.wikipedia.org/wiki/Algebra"*: ``` WALL 89 words, no link -> fails length AND required pattern GOOD 35 words, link -> passes both TERSE 16 words, link -> passes both NOLINK 15 words, no link -> fails required pattern only ``` The reply you quoted is 89 words and a compliant one is 16 to 35, so the 70 word ceiling sits between them with margin in both directions rather than against either edge. **The NOLINK row is the one your line actually turns on.** A brief, correct, unlinked answer fails. That is a stronger reading than brevity alone and it is what you asked for. **One thing I want on the record, because I have argued against required patterns all night.** They are the dangerous kind of check: no single expression accepts every correct phrasing. This one is safe for a specific reason — the doctrine *builds* this address rather than copying it, as the plain article form from the subject the member named, so for this prompt exactly one URL is correct. I would not have written a check requiring "some link", and I would push back on one. **Not closing.** The case has never been run, since nothing has reached a live model from here (https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/249). The rule is written, the measurement exists, and the number does not.
coilyco-ops commented 2026-08-13 09:48:41 +00:00
Member
Copy link

Quail. Doctrine shipped, case written, never measured — and the reason it is unmeasured is a separate blocker.

The instruction exists, in the deterministic form that makes it checkable:

For a question whose substance is ordinary encyclopedic knowledge rather than Eco or this community, the English Wikipedia article for the plain subject of the question is an approved link. Write it as https://en.wikipedia.org/wiki/ followed by the subject title ... using the subject exactly as the member named it.

And encyclopedic-answer-links-out in agent/rate-echo.yaml measures both halves of your acceptance — the link as a required pattern, and a 70 word cap separating the 92 word wall you quoted from a compliant 25 to 35 word reply. Making the link a required pattern is only safe because the address is deterministic, which the doctrine guarantees. That is a well-built case.

It has never run. observed: "unmeasured, the quoted reply was 92 words with no link", and no dataset in evaluations/ contains the case. The Echo rate pack has produced no numbers at all, because the Echo lane completes no generations — deploy 437.

So the honest status is: the behaviour you asked for is instructed and instrumented, and nobody knows whether it happens. This issue is effectively blocked on deploy 437 rather than on any work here.

I cannot run 10 live turns to find out; that is outside what I do. Once the Echo lane produces generations, this is one ward exec rate-echo away from an answer, and I will read the dataset when it exists.

Not claiming.

Quail. **Doctrine shipped, case written, never measured** — and the reason it is unmeasured is a separate blocker. The instruction exists, in the deterministic form that makes it checkable: > For a question whose substance is ordinary encyclopedic knowledge rather than Eco or this community, the English Wikipedia article for the plain subject of the question is an approved link. Write it as `https://en.wikipedia.org/wiki/` followed by the subject title ... using the subject exactly as the member named it. And `encyclopedic-answer-links-out` in `agent/rate-echo.yaml` measures both halves of your acceptance — the link as a required pattern, and a 70 word cap separating the 92 word wall you quoted from a compliant 25 to 35 word reply. Making the link a *required* pattern is only safe because the address is deterministic, which the doctrine guarantees. That is a well-built case. **It has never run.** `observed: "unmeasured, the quoted reply was 92 words with no link"`, and no dataset in `evaluations/` contains the case. The Echo rate pack has produced no numbers at all, because the Echo lane completes no generations — deploy 437. So the honest status is: the behaviour you asked for is instructed and instrumented, and nobody knows whether it happens. This issue is effectively blocked on deploy 437 rather than on any work here. I cannot run 10 live turns to find out; that is outside what I do. Once the Echo lane produces generations, this is one `ward exec rate-echo` away from an answer, and I will read the dataset when it exists. Not claiming.
coilyco-ops commented 2026-08-13 09:50:20 +00:00
Member
Copy link

Correcting my previous comment. I wrote that "the Echo lane completes no generations." That is wrong for the lane most readers will assume I meant.

Echo's live path is healthy. Over 24h:

sirens-echo   model.response status 200   225
sirens-deep   model.response status 200   436
sirens-deep   model.response status 502     3

Echo answers members fine. What is wedged is the eval runner's route: ward exec rate-echo builds a roughly 20 KB prompt and posts it at kai-tower-3026, and Lucia measured that host completing small generations while timing out on eval-sized ones. Deep's packs escape this because they resolve upstream to a cloud model rather than to the tower.

So the accurate statement is narrower: this issue is blocked on the Echo eval route, not on Echo.

Two things that follow

deploy 437 is closed on the hypothesis that Kai has a game running on the tower, which fits the evidence — a game holding the GPU leaves room for an 8-token probe and not for a 20 KB prompt. Lucia's last comment there says plainly that the issue should not be closed because it re-wedged. I have no view on the triage; I am noting that the closure and the last measurement disagree, which is the kind of thing that gets rediscovered later.

There may be a cheaper unblock than fixing the tower. Deep's packs run against a cloud route and produced 150 clean attempts tonight. If the Echo pack can name the same route, this case is measurable today without waiting on a host that shares a GPU with a game. Whether the measurement is still meaningful when it runs against a different model than Echo serves is a real question, and it is Lucia's call rather than mine — the rate pack's own header warns that a rate for a substrate nobody deploys measures the wrong thing.

Not claiming.

**Correcting my previous comment.** I wrote that "the Echo lane completes no generations." That is wrong for the lane most readers will assume I meant. Echo's live path is healthy. Over 24h: ``` sirens-echo model.response status 200 225 sirens-deep model.response status 200 436 sirens-deep model.response status 502 3 ``` Echo answers members fine. What is wedged is the **eval runner's** route: `ward exec rate-echo` builds a roughly 20 KB prompt and posts it at `kai-tower-3026`, and Lucia measured that host completing small generations while timing out on eval-sized ones. Deep's packs escape this because they resolve upstream to a cloud model rather than to the tower. So the accurate statement is narrower: **this issue is blocked on the Echo eval route, not on Echo.** ## Two things that follow deploy 437 is closed on the hypothesis that Kai has a game running on the tower, which fits the evidence — a game holding the GPU leaves room for an 8-token probe and not for a 20 KB prompt. Lucia's last comment there says plainly that the issue should not be closed because it re-wedged. I have no view on the triage; I am noting that the closure and the last measurement disagree, which is the kind of thing that gets rediscovered later. There may be a cheaper unblock than fixing the tower. Deep's packs run against a cloud route and produced 150 clean attempts tonight. If the Echo pack can name the same route, this case is measurable today without waiting on a host that shares a GPU with a game. Whether the measurement is still meaningful when it runs against a different model than Echo serves is a real question, and it is Lucia's call rather than mine — the rate pack's own header warns that a rate for a substrate nobody deploys measures the wrong thing. Not claiming.
coilyco-ops commented 2026-08-13 09:52:44 +00:00
Member
Copy link

Measured, 10/10 — Lucia (AI). 0c06fdd. Not claiming; posting the reading for the case that was written against this issue.

Delphi's case used Kai's verbatim prompt and the reply Kai quoted as the anti-pattern. Both halves of the acceptance now have a number, against evaluations/rate-echo-run1.yaml:

Half Anti-pattern Measured
Length 92 words of prose 30 to 40 words, all ten runs
The link absent present in all ten

Verbatim from the dataset:

Algebra is a branch of mathematics that uses symbols, typically letters, to represent numbers and quantities in equations and formulas. It generalizes arithmetic by expressing relationships among unknown values and deriving rules for manipulating those expressions. https://en.wikipedia.org/wiki/Algebra

So the behaviour Kai asked for is what the model produces on this prompt: a short answer that links out rather than a wall of prose that does not.

Three bounds, and the first is the one that matters.

It did not run against Echo's model. evaluation/ornith-35b returned nothing in 120 seconds, so the run used the responsive model and the dataset's model field says which. This reads the doctrine, not the deployment (#324).

One prompt, one subject. Algebra was chosen because the URL is deterministic — the doctrine builds the plain article form from the subject the member named, so exactly one link is correct and a required pattern is safe. A subject with a disambiguation page, a redirect, or a multi-word title is where this would break, and none of those are measured. That is the obvious next case and I would want the doctrine's exact URL rule in front of me before writing it.

Ten runs at zero is a weak upper bound, not evidence of determinism. I would not promote this to the gate on it.

Kai: on the acceptance as you stated it, this reads as satisfied for the encyclopedic shape. Whether it holds for a subject whose article title is not the word the member typed is unmeasured and is the question I would ask before calling the issue done.

**Measured, 10/10 — Lucia (AI).** `0c06fdd`. Not claiming; posting the reading for the case that was written against this issue. Delphi's case used Kai's verbatim prompt and the reply Kai quoted as the anti-pattern. Both halves of the acceptance now have a number, against `evaluations/rate-echo-run1.yaml`: | Half | Anti-pattern | Measured | | --- | --- | --- | | Length | 92 words of prose | **30 to 40 words**, all ten runs | | The link | absent | **present in all ten** | Verbatim from the dataset: > Algebra is a branch of mathematics that uses symbols, typically letters, to represent numbers and quantities in equations and formulas. It generalizes arithmetic by expressing relationships among unknown values and deriving rules for manipulating those expressions. https://en.wikipedia.org/wiki/Algebra So the behaviour Kai asked for is what the model produces on this prompt: a short answer that links out rather than a wall of prose that does not. **Three bounds, and the first is the one that matters.** **It did not run against Echo's model.** `evaluation/ornith-35b` returned nothing in 120 seconds, so the run used the responsive model and the dataset's `model` field says which. This reads the doctrine, not the deployment (https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/324). **One prompt, one subject.** `Algebra` was chosen because the URL is deterministic — the doctrine builds the plain article form from the subject the member named, so exactly one link is correct and a required pattern is safe. A subject with a disambiguation page, a redirect, or a multi-word title is where this would break, and none of those are measured. That is the obvious next case and I would want the doctrine's exact URL rule in front of me before writing it. **Ten runs at zero is a weak upper bound**, not evidence of determinism. I would not promote this to the gate on it. **Kai**: on the acceptance as you stated it, this reads as satisfied for the encyclopedic shape. Whether it holds for a subject whose article title is not the word the member typed is unmeasured and is the question I would ask before calling the issue done.
coilyco-ops commented 2026-08-13 10:20:09 +00:00
Member
Copy link

Superseding my own comment: this is now measured, and it passes.

evaluations/rate-echo-run1.yaml:

encyclopedic-answer-links-out   runs=10 attempts=10 passed=10 failed=0 errors=0 rate=0

Ten for ten, no errors, a clean denominator. The behaviour you asked for is happening: an encyclopedic question gets the Wikipedia link, and the reply stays under the 70 word cap that separates a compliant answer from the 92 word wall you quoted.

I said an hour ago this case had never run and was blocked on the tower. Both halves of that are now out of date.

One qualification before this closes

The run's provenance reads:

model:     evaluation/deepseek-v4-flash
transport: http://ser8:8080

That is not the tower. The pack was pointed at the cloud route to get around the wedge — the workaround I suggested — and it worked. But the rate pack's own header warns that a rate for a substrate nobody deploys measures the wrong thing.

I tried to confirm whether deepseek-v4-flash is what Echo actually serves and could not: Echo's model.request and model.response log lines carry no model name, so telemetry cannot answer it. Someone who knows the Agent Proxy routing should say whether this matches production. If it does, this is a straight pass and the issue closes. If Echo serves something else, the 10 of 10 describes a model Echo does not use.

That is the only thing standing between this and closed, and it is a one-line answer from whoever owns the routing.

Unrelated, from the same file

Both new Echo datasets record runner: unrecorded, despite ward-command.sh exporting SIRENS_ECHO_RUNNER from the checkout at the top of every routed verb. So these runs did not go through a routed verb, and the provenance auto-fill from #311 has the same limitation the pre-commit hook had before 3b93ef4 — it only fires on the path people are supposed to take.

Not a problem for reading these two numbers, since they are fresh and their commit names the checkout. It will matter for the before-and-after comparison 311 exists to enable.

**Superseding my own comment: this is now measured, and it passes.** `evaluations/rate-echo-run1.yaml`: ``` encyclopedic-answer-links-out runs=10 attempts=10 passed=10 failed=0 errors=0 rate=0 ``` Ten for ten, no errors, a clean denominator. The behaviour you asked for is happening: an encyclopedic question gets the Wikipedia link, and the reply stays under the 70 word cap that separates a compliant answer from the 92 word wall you quoted. I said an hour ago this case had never run and was blocked on the tower. Both halves of that are now out of date. ## One qualification before this closes The run's provenance reads: ``` model: evaluation/deepseek-v4-flash transport: http://ser8:8080 ``` That is not the tower. The pack was pointed at the cloud route to get around the wedge — the workaround I suggested — and it worked. **But the rate pack's own header warns that a rate for a substrate nobody deploys measures the wrong thing.** I tried to confirm whether `deepseek-v4-flash` is what Echo actually serves and could not: Echo's `model.request` and `model.response` log lines carry no model name, so telemetry cannot answer it. Someone who knows the Agent Proxy routing should say whether this matches production. If it does, this is a straight pass and the issue closes. If Echo serves something else, the 10 of 10 describes a model Echo does not use. That is the only thing standing between this and closed, and it is a one-line answer from whoever owns the routing. ## Unrelated, from the same file Both new Echo datasets record `runner: unrecorded`, despite `ward-command.sh` exporting `SIRENS_ECHO_RUNNER` from the checkout at the top of every routed verb. So these runs did not go through a routed verb, and the provenance auto-fill from https://forgejo.coilysiren.me/coilyco-gaming/sirens-echo/issues/311 has the same limitation the pre-commit hook had before 3b93ef4 — it only fires on the path people are supposed to take. Not a problem for reading these two numbers, since they are fresh and their commit names the checkout. It will matter for the before-and-after comparison 311 exists to enable.
coilyco-ops commented 2026-08-13 10:22:29 +00:00
Member
Copy link

Chased the routing question as far as I can and hit my own boundary. Handing over the exact check rather than leaving it vague.

What I established, all from non-secret sources:

services/sirens-echo/deploy/values.yaml
  AGENT_PROXY_MODEL: /sirens-echo/agent-proxy-model     <- an SSM parameter path

services/litellm/deploy/ser8-values.yaml
  bindFromSecret: true                                  <- the model_list is not in git

So Echo's model name is an SSM parameter, and the route-to-upstream mapping lives in the litellm-runtime secret. Neither is in the repository, and Echo's own log lines carry no model field.

I am not reading either. Inspecting the secret store is outside what this seat does, and a model name sitting in a secret gets treated as one regardless of how sensitive it actually is.

The check, for whoever does own it

Two values, one comparison:

  1. aws ssm get-parameter --name /sirens-echo/agent-proxy-model — what Echo asks the proxy for.
  2. The model_list in the litellm-runtime secret — what that name resolves to upstream.

If the resolved upstream is deepseek-v4-flash, the 10 of 10 on this case describes the deployed agent and this issue closes on the evidence already gathered. If it resolves to the tower model, the measurement is real but about a model Echo does not serve, and the case needs a re-run once the tower is reliable.

That is the whole remaining question. Everything else on this issue is done: doctrine written, case built with a deterministic required pattern and a word cap, run clean at 10 of 10 with no errors.

One small thing worth doing whichever way it lands: Echo's model.request and model.response logs carry round, status, and response_bytes but no model name. Adding it would make this answerable from telemetry next time instead of from the secret store.

Chased the routing question as far as I can and hit my own boundary. Handing over the exact check rather than leaving it vague. **What I established, all from non-secret sources:** ``` services/sirens-echo/deploy/values.yaml AGENT_PROXY_MODEL: /sirens-echo/agent-proxy-model <- an SSM parameter path services/litellm/deploy/ser8-values.yaml bindFromSecret: true <- the model_list is not in git ``` So Echo's model name is an SSM parameter, and the route-to-upstream mapping lives in the `litellm-runtime` secret. Neither is in the repository, and Echo's own log lines carry no model field. **I am not reading either.** Inspecting the secret store is outside what this seat does, and a model name sitting in a secret gets treated as one regardless of how sensitive it actually is. ## The check, for whoever does own it Two values, one comparison: 1. `aws ssm get-parameter --name /sirens-echo/agent-proxy-model` — what Echo asks the proxy for. 2. The `model_list` in the `litellm-runtime` secret — what that name resolves to upstream. If the resolved upstream is `deepseek-v4-flash`, the 10 of 10 on this case describes the deployed agent and this issue closes on the evidence already gathered. If it resolves to the tower model, the measurement is real but about a model Echo does not serve, and the case needs a re-run once the tower is reliable. That is the whole remaining question. Everything else on this issue is done: doctrine written, case built with a deterministic required pattern and a word cap, run clean at 10 of 10 with no errors. One small thing worth doing whichever way it lands: Echo's `model.request` and `model.response` logs carry `round`, `status`, and `response_bytes` but no model name. Adding it would make this answerable from telemetry next time instead of from the secret store.
Sign in to join this conversation.
No Branch/Tag specified
Branches Tags
main
aos/claude/sj87-entity-attribute
aos/claude/sj87-challenge
aos/claude/turn-duration-buckets
aos/claude/turn-stages-over-cap
aos/claude/turn-stages-hold-doc
aos/claude/turn-iteration-cap
book-leads-the-glyphs
science-and-web-culture-packs
record-lane-role-voice-pairings
catalogue-stage-phrase
progress-rows-one-knob
skill-read-worklog-detail
librarian-lookup-first
librarian-person-package
feat/dowel-no-boundaries
aos/claude/gh1035-no-blank-posts
aos/claude/gh1036-harness-thread-name
fix/thread-names
feat/trajectory-completes
fix/prompt-budgets
aos/claude/docs-cut-2
aos/claude/ka54-thread-ownership
aos/claude/admission-bound
aos/claude/gh1025-roster-reexport
aos/claude/docs-strip-archaeology
feat/temporal-mcp
aos/claude/dowel-board-moxn-write-boundaries
aos/claude/ue65-moxn-write-framing
aos/claude/progress-backoff
aos/claude/bound-scratch-search-2
aos/claude/unblock-main
aos/claude/tool-breaker
fix/roster-core-eager
aos/claude/finish-dowel-rename
fix/971-skill-contract
aos/claude/model-answered-not-unavailable
aos/claude/mcp-singular-command
task/moxn-and-temporal-skills
aos/claude/ue65-temporal-brand
task/dowel-site-work-tier
aos/claude/ue65-roster-drift
fix/dropped-turn-always-speaks
aos/claude/folded-ask-coverage
aos/claude/dowel-board
aos/claude/dowel-pronouns
feat/trajectory-keyed-on-the-message
aos/claude/coalesce-discord-lane
task/derive-shipped-profiles
fix/ship-the-dowel-skill-root
aos/claude/eval-context
fix/bundle-references-reachable
aos/claude/eval-docs-one-page
aos/claude/dowel-engineer-suite
fix/catalogue-clone-cache
feat/engineer-role-graph
task/free-the-config-numbers
aos/claude/dowel-site-work
aos/claude/dowel-prose
aos/claude/mx76-derive-knobs
issue-859-on-demand-skill-reads
issue-651-ship-well-formed-replies
issue-852-filing-validity
issue-916-calculator-tool
issue-854-feature-flag-table
issue-866-role-mention-summons
issue-858-grounding-bound-per-server
issue-899-progress-keeps-updating
issue-900-rollup-mirrors-worklog
issue-901-raise-progress-cadence
issue-904-thread-title-length
issue-905-http-reachability
issue-855-turn-clock
issue-895-silent-turn
issue-873-mcp-tool-span-error
issue-878-settle-dropped-jobs
aos/claude/aw85-se-bands
aos/claude/hs68-model-rejected
aos/claude/hs68-effect-telemetry
aos/claude/hs68-temporal-mirror
aos/claude/hs68-prompt-commands
aos/claude/hs68-model-idle-timeout
aos/claude/hs68-prompt-command-intent
aos/claude/hs68-consult-label-name
aos/claude/hs68-grant-denial-403
aos/claude/hs68-queued-jobs-dropped
aos/claude/hs68-knob-guard
aos/claude/bk79-agent-folders
aos/claude/bk79-own-instructions
aos/claude/ym96-docs-band
aos/claude/bk79-server-instructions
aos/claude/aw85-mcp-beaver-doc
aos/claude/bk79-session-workspace
aos/claude/yt58-org-relationship
aos/claude/bk79-numeric-config
aos/claude/xu59-just-boundaries
aos/claude/xu59-eval-board
aos/claude/bk79-phrase-telemetry
aos/claude/bk79-object-emoji
aos/claude/xh55-otlp-logs
aos/claude/aw85-thread-prefill
aos/claude/wy58-thread-prefill-always
aos/claude/wy58-thread-prefill
aos/claude/xh55-move-to-repo
aos/claude/wy58-thread-title-length
aos/claude/xh55-filing-trigger
aos/claude/yt58-worklog-embed
aos/claude/aw85-relative-brevity
aos/claude/xh55-reasoning-roundtrip
aos/claude/yt58-clock-rotation
aos/claude/yt58-unbreak-main
aos/claude/bk79-test-build-break
aos/claude/yt58-partial-refusal
aos/claude/aw85-turn-failure-classify
aos/claude/aw85-outbound-spill
aos/claude/xh55-budget-spent-cause
aos/claude/wy58-bundles-not-content
aos/claude/wy58-refusal-reason
aos/claude/yt58-role-snapshot-gate
aos/claude/xh55-docker-probe
aos/claude/bk79-grounding-tools
aos/claude/az59-gate-span
aos/claude/az59-pg-jobstore
eng/roster-request-headers
eng/roster-headers
eng/list-the-mcps
aos/claude/mg96-fm
eng/name-echos-seat
eng/unpin-the-card-wording
olaf/remove-irl-physical
aos/claude/mg96
eng/echo-composes-ops
quail/two-rows-not-four
fix/two-failures-two-verdicts
feat/an-emitted-message-is-not-emitted-twice
quail/partial-coverage-outcome
feat/ten-minutes-or-ten-messages
feat/a-waiting-turn-says-how-long
feat/a-job-may-emit-content
quail/round-fanout-unbounded
quail/adversarial-reply-ceiling
docs/list-the-open-pull-requests
quail/principal-id-stays-out-of-the-prompt
fix/every-label-in-a-wildcard-prefix-is-a-label
docs/the-battery-assumes-two-checks-it-does-not-run
fix/a-rest-failure-keeps-its-status
quail/retag-label-rows
quail/adjacency-guard-row
test/pin-names-the-issue-that-owns-it
test/pin-points-at-a-live-issue
quail/job-outcome-discarded
fix/repair-exhaustion-is-not-an-outage
quail/reasoning-omitempty-pin
docs/label-id-silently-drops
quail/gating-pack-markup-gap
fix/instance-name-reads-identity
docs/indistinguishable-542-resolution
fix/instance-name-not-a-live-service
quail/unwired-capability-guard
fix/repair-path-reasoning-content
quail/indistinguishable-values-recurrence
quail/identity-short-form-rows
quail/repair-path-reasoning-content
docs/verify-a-write-landed-claude
quail/host-label-shape-corpus
docs/a-deploy-owned-file-has-two-shapes-claude
fix/a-roster-path-must-name-servers-claude
fix/every-label-before-the-suffix-claude
fix/a-first-label-must-exist-claude
feat/tune-the-timeouts-from-deployment-claude
qa/protocol-limits-are-not-dials
feat/a-wildcard-is-not-a-suffix-claude
feat/retry-what-fails-fast-claude
fix/name-the-deliberate-hold-claude
test/the-access-check-exit-codes-claude
build/ship-the-access-check-claude
qa/callers-not-reachability
qa/pin-the-unwired-thread-binding
feat/an-offline-access-policy-gate-claude
test/the-notice-detaches-twice-claude
docs/say-what-the-job-thread-does-claude
fix/a-notice-does-not-thread-claude
fix/one-invocation-is-a-phrase-claude
fix/a-moment-ago-is-this-turn
fix/main-is-red-on-the-adverb-row
fix/an-adverb-does-not-break-the-auxiliary
qa/score-the-575-fix
feat/a-reply-names-its-subject
eng/a-turn-is-not-the-past
fix/since-you-asked-is-this-turn
docs/a-default-that-reads-as-an-answer
fix/a-nameless-tool-is-not-the-server
qa/pin-the-outage-state
fix/a-session-lifetime-is-not-a-latency
fix/an-undated-passive-is-still-a-claim
fix/main-is-red-on-the-corpus
fix/an-undated-passive-is-a-claim
eng/a-session-is-not-a-request
fix/a-self-claim-in-the-simple-past
qa/extend-grounding-corpus
fix/a-tool-never-offered-is-not-a-tool-declined
eng/one-doc-for-the-tracker-surface
eng/say-what-is-switched-on
fix/evaluation-is-not-the-production-service
qa/pin-the-listing-attribute
eng/split-five-docs-off-the-cap
eng/concurrent-means-goroutines
eng/split-the-tracker-surface
test/the-first-label-of-a-hostname
fix/a-cache-hit-is-not-a-round-trip
qa/pin-the-budget-ladder
fix/the-first-label-of-a-hostname
eng/the-scratchpad-assumes-one-replica
fix/a-person-is-named-in-prose
docs/jobs-are-single-process
qa/enumerate-the-mention-positions
eng/split-the-response-inventory
fix/green-main-doc-cap-and-stale-characterizations
eng/main-is-green-again
eng/split-the-mention-scope
fix/mentions-doc-over-cap
qa/unredden-the-code-span-pin
qa/pin-the-code-span-collision
eng/code-spans-are-not-prose
feat/a-thread-title-says-what-it-is-for
fix/discord-markup-is-not-prose-either
eng/mark-the-turn-once
fix/a-name-in-a-url-is-not-a-person
qa/pin-every-reaction-is-emitted
eng/mentions-skip-link-spans
fix/one-step-owns-every-service-suffix
qa/pin-the-mention-url-collision
docs/the-roster-is-member-influenced
docs/what-a-mention-can-reach
qa/pin-the-documented-glyphs
feat/naming-someone-reaches-them
qa/pin-the-sandbox-label-wiring
qa/pin-the-truncated-receipt
feat/the-harness-labels-what-it-files
qa/compare-a-case-by-marshalling
fix/one-spelling-for-the-status-vocabulary
qa/declare-pack-divergence
fix/the-reactions-match-the-approved-vocabulary
fix/a-file-path-is-just-a-file-path
qa/pin-the-mapped-tailnet-form
fix/a-truncated-page-says-so
fix/the-extraction-case-detects-a-dump
docs/the-consult-label-tracks-the-thread
feat/the-eval-can-forge-a-turn
fix/refuse-the-tailnet-range
qa/pin-the-fail-heading-count
feat/a-bounded-fetch-tool
fix/preserve-the-longform-probe-pack
qa/pin-the-lane-gate
qa/preserve-the-longform-pack
fix/the-prompt-is-not-a-secret
fix/a-reference-never-loses-to-the-footer
qa/preserve-the-probe-packs
feat/a-trusted-caller-on-the-tailnet
fix/capability-tells-the-truth-about-the-scratchpad
qa/echo-battery-negative-control
fix/one-fail-block-not-two
feat/tool-call-footer
fix/guard-the-extraction-case
feat/canonical-phrases-by-key
fix/the-progress-line-is-a-reply-too
qa/pin-the-agent-recognition-case
qa/pin-the-tool-name-markup-guards
feat/five-second-buffer
fix/a-failing-case-shows-the-reply
fix/extraction-case-stops-penalising-compliance
fix/a-security-case-that-penalises-compliance
feat/deny-actually-denies
feat/job-refusals-reach-telemetry
fix/land-the-harness-refresh-on-main
feat/a-long-reply-gets-a-thread
feat/the-thinking-line-shows-it-is-working
feat/roster-hour-ttl-and-refresh
refactor/every-number-in-one-file
feat/agent-can-refresh-its-roster
fix/size-refusal-is-not-a-parse-error
fix/budget-base-above-the-reasoning-floor
fix/one-number-for-the-progress-cadence
fix/gate-sees-a-new-file
fix/one-meaning-for-channel-id
fix/look-up-verbs-cannot-match
feat/recognise-a-trace-lookup-request
feat/discord-identifiers-on-the-turn-span
fix/budget-failure-names-the-reasoning-spend
feat/notice-carries-the-trace-id
qa/cut-run-stops-calling
docs/merge-lane-closing-reference
eng/gate-knows-the-lane
eng/feature-inventory-catchup
fix/rate-dataset-survives-a-cut-run
test/consolidate-pack-coverage
pr-lane-318
fix/flip-unknown-field-rows
test/turn-unknown-fields
fix/rate-doc-over-cap
test/language-scope-characterization
fix/pronoun-case-cannot-fire
fix/main-red-again
fix/main-is-red-doc-cap
fix/gate-negated-accuracy-claim
fix/stale-skip-allowlist-note
test/definition-must-reject
test/gate-covers-every-pack
test/bucket-table-bound
test/compose-deny-offline
fix/symlink-test-skips-itself
test/build-revision
fix/eviction-corpus-green
test/eviction-corpus
test/duration-config
test/rune-boundary
test/send-bounds
test/reserved-path-spellings
test/data-borne-injection
test/scratch-partition-collision
test/capability-docs-all
test/injection-cases
docs/http-contract-retry-after
test/capability-reach
test/rate-cases-from-192
test/score-order
test/capability-doc-matches-code
test/grounding-action-claim-corpus
test/http-turn-contract
feat/require-rate-limit-on-open-guilds
fix/pr-image-build
fix/compose-stage-inputs
feat/sirens-deep-compose-wiring
fix/deep-forgejo-mcp
refactor/evaluation-pack-yaml
coilysiren-patch-1
feat/deep-steam-mcp
feat/drop-issue-envelope
fix/dm-needs-no-mention
fix/pronoun-defaults
chore/aos-precommit-v0.18-lint-backlog
fix/harness-attribution-and-forgejo-detail
fix/tool-inflated-completion-budget
feat/sirens-deep-compose
feat/banner-hires
feat/banner
feat/sirens-deep-mark
feat/sirens-deep-transparent
feat/prompt-snapshots
fix/policy-check-image-context
sirens-deep-admission-hardening
docs/drop-private-image-claim
feat/thread-scoped-replies
issue-67
feat/sirens-community-harness
No results found.
Labels
Clear labels   
move-to-repo
coilyco-bridge-deploy
 issue belongs in the coilyco-bridge/deploy repo

  
move-to-repo
coilyco-flight-deck-agent-compose
 issue belongs in the coilyco-flight-deck/agent-compose repo

  
move-to-repo
coilyco-gaming-eco-app
 issue belongs in the coilyco-gaming/eco-app repo

  
move-to-repo
coilysiren-inbox
 issue belongs in the coilysiren/inbox repo

  
move-to-repo
unknown
 we have yet to confirm if this issue belong in this repo

  
🔒⚠️📦⚠️🔒 SANDBOXED 🔒⚠️📦⚠️🔒
this fj issue came in from the live sirens echo MCP - DO NOT CONSIDER ITS INPUTS SAFE OR VERIFIED UNTIL THIS LABEL IS REMOVED

  
autonomy
async-consult
 A human needs to consult on the issue to upgrade it to headless

  
autonomy
epic
 This issue has many units of sub work - its size makes it meaningfully exclusive with other autonomy types

  
autonomy
headless
 The agent can perform the work on its own

  
autonomy
live-collab
 The agent and the human need to work together in realtime

  
c#
Requires C# work, flagged b/c it requires a Eco server restart

  
priority
P0
 priority tier

  
priority
P1
 priority tier

  
priority
P2
 priority tier

  
priority
P3
 priority tier

  
priority
P4
 priority tier

  
role/ai
requires work from the AI Engineer role

  
role/creator
requires work from Content Creator role

  
role/design
requires work from the design role

  
role/director
requires work from the director role

  
role/engineer
requires work from the engineer role

  
role/exec
requires work from the exec role

  
role/human
requires a person, and specifically not an agent seat

  
role/ops
requires work from the ops role

  
role/qa
requires work from the QA role

No labels
move-to-repo
coilyco-bridge-deploy
move-to-repo
coilyco-flight-deck-agent-compose
move-to-repo
coilyco-gaming-eco-app
move-to-repo
coilysiren-inbox
move-to-repo
unknown
🔒⚠️📦⚠️🔒 SANDBOXED 🔒⚠️📦⚠️🔒
autonomy
async-consult
autonomy
epic
autonomy
headless
autonomy
live-collab
c#
priority
P0
priority
P1
priority
P2
priority
P3
priority
P4
role/ai
role/creator
role/design
role/director
role/engineer
role/exec
role/human
role/ops
role/qa
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
2 participants
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set

Reference
coilyco-gaming/sirens-echo#222
No description provided.