docs(#6785): add text/template sorted map iteration to Go review guidance

[fullsend-ai-coder]

Summary

Adds a "Template map iteration" review directive to docs/contributing/go-code.md to prevent the review agent from flagging range over basic-type-keyed maps in text/template as non-deterministic. Go's text/template package guarantees sorted iteration for string, int, and float keys — a documented exception to Go's general unordered map iteration.

Related Issue

Changes

• Added a new ### Template map iteration subsection under ## Go pitfalls in docs/contributing/go-code.md
• Includes a "When reviewing PRs" directive following the established pattern of the Timeout() bool and suite-timeout directives
• Links to the official text/template documentation

Testing

• No Go production code changed — docs-only change
• Pre-commit hooks verified (trailing whitespace, end-of-file, link checks)
• Secret scan passed
• Gitlint passed

---

Closes #6785

Post-script verification

• Branch is not main/master (agent/6785-template-map-iteration-guidance)
• Secret scan passed (gitleaks — b00042354a39a430b7658aa186ab95690fdfbbb4..HEAD)
• PR body secret scan passed (gitleaks — no-git)

[github-actions]

Site preview

Preview: https://6403aa4e-site.fullsend-ai.workers.dev

Commit: 1cd5370ff93e2f5867318cca13f171e1516ff2b2

[fullsend-ai-review]

🤖 Finished Review · ✅ Success · Started 9:16 PM UTC · Completed 9:29 PM UTC

Commit: 68f351e · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $3.37

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[fullsend-ai-review]

Risk Assessment: low (1/5)

Details

Docs-only PR by a bot adding 15 lines to a single contributing guide, with no protected paths, no security-sensitive files, and exact scope match to the linked issue.

Previous run

Risk Assessment: low (1/5)

Details

Docs-only PR by a bot adding 15 lines to a single contributing guide, with no protected paths, no security-sensitive files, and exact scope match to the linked issue.

[fullsend-ai-review]

Looks good to me

Previous run

Review

Findings

Low

• [technical accuracy] docs/contributing/go-code.md:314 — The parenthetical list of basic types says "(string, int, float)" but omits unsigned integer types (uint, uint8, uint16, uint32, uint64) which are also basic types with a defined order that receive sorted iteration in text/template. The official Go documentation says "basic type with a defined order" without limiting to only string/int/float. The "When reviewing PRs" directive on line 323 repeats the incomplete list ("string, int, and float keys") without the "basic types" qualifier, making the omission more consequential — a reviewer reading this guidance might incorrectly conclude that a map[uint64]T would NOT get sorted iteration in a template.
  Remediation: Expand the parenthetical to "(string, int, uint, float)" or use the official phrasing "basic type with a defined order" to match the upstream documentation. Apply the same fix to the "When reviewing PRs" directive on line 323.

---

Labels: PR modifies only documentation under docs/contributing/

---

Next steps:

• /fs-fix — agent addresses review findings automatically
• /fs-fix <your instruction> — agent fixes with your specific guidance
• Push commits directly — review re-runs automatically on push
• /fs-fix-stop — disable automatic fix runs for this PR

[fullsend-ai-coder]

🤖 Finished Fix · ✅ Success · Started 9:31 PM UTC · Completed 9:36 PM UTC

Commit: 68f351e · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $1.03

[fullsend-ai-coder]

🔧 Fix agent — iteration 1 (bot-triggered)

Addressed 1 review finding: expanded the basic type list in both the description and the reviewer directive to include uint alongside string, int, and float, matching the upstream text/template documentation.

Fixed (1):

1. Incomplete basic type list omits uint types in template map iteration guidance (docs/contributing/go-code.md): Added uint to the parenthetical list on line 314 ("string, int, uint, float") and updated the "When reviewing PRs" directive on line 323 to include "uint" in the enumerated key types. The Go text/template documentation specifies sorted iteration for "basic type with a defined order" which includes unsigned integer types.

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-review]

🤖 Finished Review · ✅ Success · Started 9:38 PM UTC · Completed 9:50 PM UTC

Commit: 1cd5370 · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $3.14

Superseded by updated review

[fullsend-ai-retro]

🤖 Finished Retro · ✅ Success · Started 9:29 AM UTC · Completed 9:40 AM UTC

Commit: 1cd5370 · View workflow run →

Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $3.82

[fullsend-ai-retro]

Retro: PR #6786 — Add text/template sorted map iteration to Go review guidance

This was a well-executed end-to-end workflow with no novel improvement opportunities. The automated pipeline (triage → code → review → fix → re-review) completed in ~50 minutes, and the review agent caught a legitimate technical accuracy issue that the code agent missed.

Timeline

Time (UTC)	Stage	Run	Cost	Outcome
Aug 30 21:01–21:07	Triage	33335267707	—	Labeled ready-to-code
Aug 30 21:07–21:14	Code	33335550330	—	Created PR #6786
Aug 30 21:14–21:29	Review (1st)	33335851315	$3.37	1 low finding: uint omitted from basic type list
Aug 30 21:29–21:36	Fix	33336555776	$1.03	Added uint to type list
Aug 30 21:36–21:51	Review (2nd)	33336906484	$3.14	Approved, no findings
Sep 2 09:26	Human review	—	—	Approved (rh-hemartin, no comments)
Sep 2 09:28	Merged	—	—	Via merge queue

Total agent cost (review + fix): $7.54. Agents repo: fullsend-ai/agents@802bbd4fcade.

What went well

1. Review quality: The review agent caught a genuine technical accuracy gap — the code agent listed (string, int, float) as the sorted-iteration types but omitted uint, which the Go text/template docs include under "basic type with a defined order." The finding was actionable and correctly scoped.
2. Fix quality: The fix agent addressed the finding in one iteration (7 minutes, $1.03), adding uint to both the description and the reviewer directive.
3. Pipeline velocity: The full automated cycle completed in 50 minutes. One review–fix round, no oscillation.
4. Risk assessment accuracy: The review agent correctly scored this as risk 1/5 (docs-only, single file, 15 lines, bot author).

Evidence for existing issues (no new proposals needed)

• Add pre-flight scope heuristic to adjust review effort/model for trivially-scoped PRs #6891 (pre-flight scope heuristic for trivially-scoped PRs): This PR adds another data point. The review agent ran at effort: high on Opus for a 15-line docs-only change it rated 1/5 risk, costing $3.37 + $3.14 = $6.51 across two reviews. A lower effort or lighter model for PRs matching this profile would reduce cost without quality loss — the substantive finding (missing uint) is a straightforward completeness check that a lighter model would likely still catch.
• Track autonomy readiness: review agent fully covered human review on docs-only mechanical path update #5289 (autonomy readiness for docs-only changes): The human reviewer approved with no comments 2.5 days after the agent pipeline completed, adding zero analytical delta. This is another evidence point for the pattern tracked in that issue: review agent fully covers human review on docs-only changes.