coilyco-flight-deck/voice-flow-learning-loop
Archived
1
0
Fork 0
Code Issues 11 Pull requests Projects Releases Packages Wiki Activity Actions

Loop 2 layer 6: phrase-shape extractors #5

New issue
Open
opened 2026-05-23 20:55:37 +00:00 by coilysiren · 0 comments
coilysiren commented 2026-05-23 20:55:37 +00:00
Owner
Copy link

Originally filed by @coilysiren on 2026-05-20T12:01:15Z - https://github.com/coilysiren/voice-flow-learning-loop/issues/12

Context

Layer 6 of the 8-layer build path for Loop 2 documented in README.md.

Goal

"Phrase" isn't one shape. Three categories deserve separate extractors and weightings:

  • Path-shaped: slash-separated, often start with / or ~/, no spaces. Examples: /Users/kai/projects/coilysiren/..., ~/.repo-recall/dispatch/, /coilysiren/kai-server/tailnet-fqdn.
  • Command-shaped: start with a known CLI verb (coily, gh, kubectl, tailscale, aws), or with - / -- flags. Examples: coily ops gh issue create, --commit-scope=/Users/kai/projects/agentic-os-kai.
  • Prose-shaped: natural-language n-grams, sentence fragments. Examples: **Problem**\n\n, issue-template openers, conversational filler.

Each shape gets its own extractor with shape-aware boundary detection (paths break on whitespace; commands include flag clusters as one unit; prose extracts n-grams). Each shape gets its own ranking weight, tunable.

Done when

  • Three extractors exist as separate functions in scripts/find-snippet-candidates.py (or wherever the layer 2/3 logic lives).
  • Candidate output is grouped by shape.
  • Per-shape weighting is tunable via config or CLI flags.

Depends on: layer 5. Unblocks: cleaner candidate set for layer 7.

_Originally filed by @coilysiren on 2026-05-20T12:01:15Z - [https://github.com/coilysiren/voice-flow-learning-loop/issues/12](https://github.com/coilysiren/voice-flow-learning-loop/issues/12)_ **Context** Layer 6 of the 8-layer build path for Loop 2 documented in [README.md](../blob/main/README.md). **Goal** "Phrase" isn't one shape. Three categories deserve separate extractors and weightings: - **Path-shaped:** slash-separated, often start with `/` or `~/`, no spaces. Examples: `/Users/kai/projects/coilysiren/...`, `~/.repo-recall/dispatch/`, `/coilysiren/kai-server/tailnet-fqdn`. - **Command-shaped:** start with a known CLI verb (`coily`, `gh`, `kubectl`, `tailscale`, `aws`), or with `-` / `--` flags. Examples: `coily ops gh issue create`, `--commit-scope=/Users/kai/projects/agentic-os-kai`. - **Prose-shaped:** natural-language n-grams, sentence fragments. Examples: `**Problem**\n\n`, issue-template openers, conversational filler. Each shape gets its own extractor with shape-aware boundary detection (paths break on whitespace; commands include flag clusters as one unit; prose extracts n-grams). Each shape gets its own ranking weight, tunable. **Done when** - Three extractors exist as separate functions in `scripts/find-snippet-candidates.py` (or wherever the layer 2/3 logic lives). - Candidate output is grouped by shape. - Per-shape weighting is tunable via config or CLI flags. **Depends on:** layer 5. **Unblocks:** cleaner candidate set for layer 7.
coilysiren added
P4
 and removed
P3
 labels 2026-05-31 07:01:25 +00:00
Commenting is not possible because the repository is archived.
No Branch/Tag specified
Branches Tags
main
No results found.
Labels
Clear labels
  
P0
now / blocking / urgent

  
P1
next / important

  
P2
backlog / will do

  
P3
someday / low

  
P4
icebox / frozen / reopen if it becomes real

No labels
P0
P1
P2
P3
P4
Milestone
Clear milestone
No items
No milestone
Projects
Clear projects
No items
No project
Assignees
Clear assignees
No assignees
1 participant
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-flight-deck/voice-flow-learning-loop#5
No description provided.