skill-authoring §3: capability-negative claims rot the worst - #58
Merged
Conversation
'No such flag' / 'only works interactively' / 'the API can't do X' are version-scoped observations that read as timeless rules. A stale positive fails loud when followed; a stale negative fails silent — it steers every later session away from a capability that now exists and nothing ever exercises it to expose the rot. Incident shape: a playbook's no-flag claim was actively wrong at the tool's current version and had been routing sessions into a degraded path. Rule: pin capability-negatives to the observed version; re-verify on version change before repeating. Ships unprobed with Provenance per the covenant.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Adds one rule to skill-authoring §3 (provenance and decay): capability-negative claims rot the worst — "no such flag" / "only works interactively" / "the API can't do X" are version-scoped observations; pin each to the version it was observed on and re-verify (one
--helpor probe call) when the tool's version changes, before repeating it.Why
The section already date-stamps volatile facts, but negatives have an asymmetric failure mode the existing rule doesn't name: a stale positive fails loud the first time someone follows it, while a stale negative fails silent — it steers every later session away from a capability that now exists, and nothing ever exercises it to expose the rot. Incident shape: a playbook's "model switching only works interactively; no flag" was actively wrong at the tool's current version and had been routing sessions into a degraded interactive-only path until a review pass re-probed the binary.
Covenant status
unprobed, private incident cited as shape per the README covenant's second branch; Provenance names the unrun probe (re-run recorded capability-negatives on each version change, count flips) as debt.Checks
Single-file diff (+23). Added-lines unicode sweep: only house-style chars (§ — ❌).
python3 .github/checks.py: all checks passed.