fix(skillgen): clarify empty results in honesty rules - #2680
Conversation
…2672) When an empty result is returned, the agent shouldn't immediately assume the nodes are absent from the codebase, but absent from the graph, since things like skipped sensitive files or structurally skipped content might be the root cause. This change adds two new Honesty Rules to explicitly instruct the agent on how to handle empty results and broken commands.
|
Here is the implementation addressing #2672. The new rules were added to the core fragments ( |
![graphify-labs[bot]](https://avatars.githubusercontent.com/in/4356270?s=60&v=4){kind=small}
There was a problem hiding this comment.
Graphify reviewed this change.
Looks safe to merge — no coupling regressions and no blocking issues, checked against the code graph (not a self-assessment).
Graphify review — findings
This PR appends a two-item "honesty rules" block to the end of the graphify skill definition files across many agent variants (skill-agents.md, skill-aider.md, skill-amp.md, skill-claw.md, skill-codex.md, skill-copilot.md, skill-devin.md, skill-droid.md, skill-kilo.md, and others), along with the corresponding generator fragments/expected test fixtures under tools/skillgen. The added guidance instructs against treating empty traversal/query results as corpus-wide absence claims and against trusting empty output from commands that may have silently failed. The change appears to be documentation/prompt content propagated identically across variants, plus supporting skillgen updates.
No blocking issues surfaced.
Analysis details — impact, health, verification
Impact & health
Graphify review
Impact — 1096 functions depend on the 1032 functions this change touches.
Health — grade A; 5 existing hotspot(s) in the area this change touches (pre-existing, not introduced here):
render()— 13 callers, 5 callees (high)audit_coverage()— 8 callers, 6 callees (high)main()— 3 callers, 11 callees (medium)monolith_roundtrip()— 3 callers, 5 callees (medium)test_audit_catches_a_dropped_non_allowlisted_heading()— 0 callers, 6 callees (medium)
Verification — 1096 functions in the blast radius were not formally verified this run (proofs are advisory here).
Gate & verification
graphify gate
PASS — objectively clean (no health regressions, tests not run — proofs not run this pass (advisory)). Grounded, not self-assessed.
Advisory (not blocking):
- verification_scope: 1096 function(s) in the blast radius were not formally verified this run
|
Hey @safishamsi ! can this pr be reviewed |
1 participant
Fixes #2672