docs(#608): migrate off .fullsend/customized/skills/ per ADR 0064 by fullsend-ai-coder[bot] · Pull Request #609 · fullsend-ai/agents

fullsend-ai-coder · 2026-07-31T18:09:57Z

Summary

Remove all references to the deprecated .fullsend/customized/skills/ overlay directory from AGENTS.md, docs/review.md, and docs/triage.md per ADR 0064.
Replace org-level skill override guidance with harness base: composition pattern (harness skills: entries with paths/URLs, composed via base:).
Add a concrete worked example in docs/triage.md showing how to override issue-labels via a custom triage harness with base: composition.

Changes

AGENTS.md § 7 (Skill resolution): Updated the resolution chain from three levels (repo → org overlay → platform) to two (repo → platform). Added a paragraph explaining how to override skills using harness base: composition, with a cross-reference to the existing worked example in docs/code.md.

docs/review.md: Replaced the customized/skills/ org-level override instruction with a reference to base: composition.

docs/triage.md: Same replacement as review.md, plus a new YAML example showing how to create a .fullsend/triage.yaml with base: pointing to the upstream harness and a local skills: entry.

Testing

Verified zero remaining customized/skills references via grep -r 'customized/skills'
Secret scan passed on all changed files
Gitlint passed on the commit message
Changes are docs-only — no runtime behavior affected

Closes #608

Post-script verification

Branch is not main/master (agent/608-migrate-off-customized-skills)
Secret scan passed (gitleaks — a0fec953cd3c8b877af9821551b503df88d394d8..HEAD)
PR body secret scan passed (gitleaks — no-git)
Pre-commit hooks passed (authoritative run on runner)
Tests ran inside sandbox

Remove references to the deprecated org-level overlay directory (.fullsend/customized/skills/) from AGENTS.md, docs/review.md, and docs/triage.md. Replace with guidance to use harness base: composition for skill overrides. AGENTS.md § 7 now describes two-level resolution (repo-level and platform-level) and points to harness base: composition for org-level overrides. docs/review.md and docs/triage.md replace the customized/ overlay instructions with base: composition guidance, and docs/triage.md adds a concrete worked example showing how to override issue-labels via a custom triage harness. Note: pre-commit could not run (network error fetching hooks in sandbox). The post-script runs pre-commit authoritatively on the runner. Closes #608

fullsend-ai-review · 2026-07-31T18:11:45Z

🤖 Finished Review · ✅ Success · Started 6:11 PM UTC · Completed 6:25 PM UTC
Commit: 52f65d5 · View workflow run →

fullsend-ai-review · 2026-07-31T18:25:29Z

Review

Findings

Medium

[protected-path] AGENTS.md — This PR modifies AGENTS.md, which is a protected governance file. The linked issue (docs: migrate off .fullsend/customized/skills/ per ADR 0064 #608) and PR description explain the rationale (migrating off deprecated .fullsend/customized/skills/ overlay per ADR 0064). Human approval is always required for protected-path changes, regardless of context.

Low

[terminology-consistency] AGENTS.md:83 — Changed "Skills declared in agent frontmatter" to "Skills listed in harness". The verb change from "declared" to "listed" is a minor stylistic preference with no impact on clarity.

Labels: PR is a docs-only change updating AGENTS.md and docs/ files.

fullsend-ai-review

See the review comment for full details.

fullsend-ai-review · 2026-07-31T18:25:30Z

-in frontmatter is valid even if no matching directory exists in this repo.
-Do not treat missing local skill directories as bugs without first
-verifying the skill does not exist at org or platform level.
+Skills listed in harness `skills:` arrays are resolved at runtime from


[low] terminology-consistency

Changed 'Skills declared in agent frontmatter' to 'Skills listed in harness'. The verb change from 'declared' to 'listed' is a minor stylistic preference with no impact on clarity.

ralphbean

I think this needs a couple changes before we can merge. See inline comments.

ralphbean · 2026-07-31T18:50:52Z

+
+To override an upstream skill at the org or repo level, create a custom
+harness with `base:` composition pointing to the upstream harness and
+include the replacement skill in the `skills:` array. See


[moderate] The code.md#how-to-configure example overrides the sandbox image:, not skills: — someone clicking through for skills guidance lands somewhere else. Could we drop a skills-specific example inline here instead, like triage.md does, rather than cross-linking?

ralphbean · 2026-07-31T18:50:52Z

-the upstream default — no other configuration needed.
+To override at the org level, create a custom harness with `base:`
+composition and include the replacement skill in the `skills:` array (see
+[Custom sandbox image — How to configure](code.md#how-to-configure)).


[moderate] Same as the AGENTS.md comment — this points at code.md's image-override example for skills guidance. Worth putting the example inline here too?

ralphbean · 2026-07-31T18:50:52Z

+valid even if no matching directory exists in this repo. Do not treat
+missing local skill directories as bugs without first verifying the skill
+does not exist at platform level.
+


[important] The resolution chain right above this just dropped org-level entirely — repo and platform only. Is "org level" still the right word here? This override is scoped to whatever repo holds the .fullsend/*.yaml file, not the whole org.

ralphbean · 2026-07-31T18:50:52Z

-You can also overload it at the org level in your `.fullsend` config repo at
-`customized/skills/issue-labels/SKILL.md`. At runtime, your version replaces
-the upstream default — no other configuration needed.
+To override at the org level, create a custom harness with `base:`


[important] Same "org level" question as the AGENTS.md comment — this override is repo-scoped, right?

ralphbean · 2026-07-31T18:50:52Z

-You can also overload it at the org level in your `.fullsend` config repo at
-`customized/skills/issue-labels/SKILL.md`. At runtime, your version replaces
-the upstream default — no other configuration needed.
+To override at the org level, create a custom harness with `base:`


[important] Same "org level" question as the AGENTS.md comment.

ralphbean · 2026-07-31T18:50:57Z

/fs-fix address the review findings

fullsend-ai-coder · 2026-07-31T18:52:38Z

🤖 Finished Fix · ✅ Success · Started 6:52 PM UTC · Completed 6:58 PM UTC
Commit: 52f65d5 · View workflow run →

- Remove stale "org level" / "org or repo level" terminology from AGENTS.md, docs/review.md, and docs/triage.md — the resolution chain now has only repo-level and platform-level, and the base: override is scoped to whichever repo holds the .fullsend/*.yaml file. - Replace misleading cross-link to code.md#how-to-configure (which shows an image override, not a skills override) with inline YAML examples in AGENTS.md and docs/review.md, matching the pattern already used in docs/triage.md. - Apply reviewer's suggested wording ("To override explicitly") in docs/review.md and docs/triage.md. Addresses review feedback on #609

fullsend-ai-coder · 2026-07-31T18:58:24Z

🔧 Fix agent — iteration 1 (human-triggered)

Addressed all 6 review findings (3 important, 2 moderate, 1 suggestion) by removing stale 'org level' terminology, replacing misleading code.md cross-links with inline skills-override YAML examples, and applying the reviewer's suggested wording.

Fixed (4):

AGENTS.md: 'org or repo level' terminology inconsistent with new resolution chain (AGENTS.md): Removed 'at the org or repo level' qualifier — the override is repo-scoped via .fullsend/*.yaml, and the resolution chain now only has repo-level and platform-level.
AGENTS.md: cross-link to code.md#how-to-configure points to image-override example, not skills (AGENTS.md): Replaced the generic 'for a worked example' cross-link with an inline YAML example showing base: composition with a skills: override (matching the pattern already in triage.md). Retained the cross-link narrowly for SHA/sha256sum values.
docs/review.md: 'org level' terminology and misleading cross-link (docs/review.md): Applied reviewer's suggested wording ('To override explicitly'), removed the cross-link to code.md#how-to-configure, and added an inline YAML example with a narrower cross-link for SHA/sha256sum values.
docs/triage.md: 'org level' terminology inconsistent with new resolution chain (docs/triage.md): Changed 'To override at the org level' to 'To override explicitly' for consistency with the other files and the new resolution chain.

Tests: passed

Next steps:

/fs-review — request a re-review of the changes
/fs-fix <your instruction> — run another fix pass with specific guidance
Push commits directly — review re-runs automatically on push
_{Updated by fullsend fix agent}

fullsend-ai-coder Bot requested a review from a team as a code owner July 31, 2026 18:09

fullsend-ai-coder Bot added the ready-for-review label Jul 31, 2026

fullsend-ai-coder Bot assigned ralphbean Jul 31, 2026

fullsend-ai-review Bot reviewed Jul 31, 2026

View reviewed changes

fullsend-ai-review Bot added requires-manual-review Review requires human judgment documentation Improvements or additions to documentation labels Jul 31, 2026

ralphbean reviewed Jul 31, 2026

View reviewed changes

Comment thread docs/review.md Outdated

ralphbean requested changes Jul 31, 2026

View reviewed changes

Uh oh!

Conversation

fullsend-ai-coder Bot commented Jul 31, 2026

Summary

Changes

Testing

Post-script verification

Uh oh!

fullsend-ai-review Bot commented Jul 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

fullsend-ai-review Bot commented Jul 31, 2026

Review

Findings

Medium

Low

Uh oh!

fullsend-ai-review Bot left a comment

Choose a reason for hiding this comment

Uh oh!

fullsend-ai-review Bot Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

ralphbean left a comment

Choose a reason for hiding this comment

Uh oh!

ralphbean Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

ralphbean Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

ralphbean Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

ralphbean Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

ralphbean Jul 31, 2026

Choose a reason for hiding this comment

Uh oh!

ralphbean commented Jul 31, 2026

Uh oh!

fullsend-ai-coder Bot commented Jul 31, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

fullsend-ai-coder Bot commented Jul 31, 2026

🔧 Fix agent — iteration 1 (human-triggered)

Tests: passed

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

fullsend-ai-review Bot commented Jul 31, 2026 •

edited

Loading

fullsend-ai-coder Bot commented Jul 31, 2026 •

edited

Loading