briefs: expand to 7 roles + add failure-analyst#7
Merged
Conversation
Patton's request, KI7MT-approved. Six existing briefs (~20 lines each) expanded to ~80-120 lines using a consistent seven-section skeleton. New seventh role failure-analyst added. Skeleton (every brief) 1. One-paragraph role + disposition 2. Strengths concrete capabilities, not adjectives 3. Avoids what the agent declines, even if asked 4. Inputs what to send, in what format 5. Outputs what to expect back, in what shape 6. Hand-offs who receives output and what they need 7. When to use right time vs wrong time Brief line counts (before -> after) architect.md 20 -> 91 implementer.md 20 -> 87 operator.md 18 -> 97 ops.md 19 -> 96 reviewer.md 22 -> 88 tester.md 20 -> 96 failure-analyst.md 0 -> 115 (new) total 119 -> 670 Why a failure-analyst slot Distinct from reviewer (reviewer reads diffs and reports; failure- analyst writes throwaway probes to falsify claims, then reports what survived) and from tester (tester writes coverage; failure-analyst writes attacks). The combination - does not write production code, writes adversarial tests, owns the failure-mode taxonomy - was not captured in any of the six existing briefs. examples/README.md Rewritten to introduce the seven roles, include an ASCII pipeline diagram, document the seven-section skeleton, and walk through copying the bundle into the runtime briefs directory. README.md (root) One-paragraph pointer at the brief-format section directing readers to examples/briefs/ and examples/README.md. Closes the discoverability gap Patton flagged. Pending: Watson voice review before merge (Patton's explicit ask). Tests + lint still clean (159 Python + 7 Go, ruff + golangci-lint both 0 issues). No code changes - content only.
KI7MT
added a commit
that referenced
this pull request
May 10, 2026
Patton's expansion pass + failure-analyst slot (PR #7, merged) is the first user-facing content beat since the v0.3 series started. Minor bump fits — meaningful new surface for operators copying the bundle, not just internal CI/test infra.
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
1 participant
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.
Summary
Patton's request (KI7MT-approved via inbox UI). Expands the six existing example briefs from sketches (~20 lines each) to full role definitions (~80–120 lines each), and adds a seventh role —
failure-analyst— that closes a real gap in the example set.Type of change
What changed
Every brief uses the same seven-section skeleton: role + disposition, strengths, avoids, inputs, outputs, hand-offs, when to use.
Why a failure-analyst slot
Distinct from reviewer (reviewer reads diffs and reports; failure-analyst writes throwaway probes to falsify claims, then reports what survived) and from tester (tester writes coverage; failure-analyst writes attacks). The combination — does not write production code, writes adversarial tests, owns the failure-mode taxonomy — wasn't captured in any of the existing six.
Pending: Watson voice review
Patton's explicit ask was to run the bundle past Watson for consistency-of-voice review before committing. CI runs on this PR; merge held until Watson signs off via inbox.
Checklist