docs: add crawl-walk-run-fly adoption guide - #4905
Conversation
|
🤖 Finished Review · ✅ Success · Started 11:18 PM UTC · Completed 11:31 PM UTC |
PR Summary by QodoDocs: add crawl-walk-run-fly incremental adoption guide
AI Description
Diagram
High-Level Assessment
Files changed (2)
|
Code Review by Qodo
1.
|
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
ReviewFindingsHigh
Low
Next steps:
Previous runRe-review at Technical claims re-verified by correctness sub-agent: code agent scoping (bugs, docs, performance), fix agent bot-authored PR auto-trigger, retro agent post-script issue filing, CODEOWNERS-based auto-merge mechanism, Remaining notes (non-blocking)1. No explanation of how to selectively enable agents — lowFile: Each stage says to "enable" specific agents without explaining the mechanism. A brief inline note or link to the 2. Guide audience partially overlaps with getting-started — lowFile: The guide explicitly targets "team leads evaluating how to roll out fullsend, and for developers who want to understand what to expect as usage evolves." Per ADR 0023, user/ is for "developers working in repositories where fullsend is active." The developer-facing half fits user/, but the team-lead-facing half aligns more with getting-started/. This is a reasonable editorial choice — noting for future information architecture consideration. Previous run (2)Re-review at Technical claims re-verified: code agent scoping (bugs, docs, performance), fix agent bot-authored PR auto-trigger, retro agent post-script issue filing, CODEOWNERS-based auto-merge mechanism — all accurate against current agent documentation. All 13 cross-reference links resolve. No secrets, injection patterns, or suspicious Unicode. Remaining notes (non-blocking)1. No explanation of how to selectively enable agents — lowFile: Each stage says to "enable" specific agents without explaining the mechanism. A brief inline note or link to the Previous runReview — approveRe-review at
The new commit also adds a note that All 13 cross-reference links resolve correctly. No secrets, injection patterns, or suspicious Unicode detected. Technical claims verified against authoritative agent documentation. Sidebar and README entries follow established formatting patterns. Remaining notes (non-blocking)1. No explanation of how to selectively enable agents — lowFile: Each stage says to "enable" specific agents without explaining the mechanism. A brief inline note or link to the 2. No linked issue — lowFile: This is a 173-line addition with no linked issue. The content clearly aligns with project goals and the contribution is well-structured. Consider linking to or filing an issue for traceability. Previous run (3)Review — commentRe-review at
All three fixes verified against the actual agent documentation. The technical accuracy of the guide is solid. One medium-severity finding from the prior review remains unaddressed. No new issues introduced. Findings1. Missing VitePress sidebar entry — mediumFile: Per AGENTS.md: "When adding a new doc under The "User Guides" sidebar section (config.ts lines 218–232) uses manual entries, not Remediation: Add 2. No explanation of how to selectively enable agents — lowFile: Each stage says to "enable" specific agents (e.g., "Enable the retro agent" in Walk) without explaining the mechanism. The guide links to 3. No linked issue — lowFile: This is a 172-line addition with no linked issue. The content clearly aligns with project goals and the contribution is well-structured. Consider linking to or filing an issue for traceability. All cross-reference links (13 total) resolve correctly. No security concerns (no secrets, injection patterns, or suspicious Unicode). Documentation formatting follows established patterns in the user guides directory. Previous runReview — commentUseful documentation contribution that provides a clear incremental adoption path. The crawl-walk-run-fly framing is well-structured and the tone is appropriate (suggestions, not requirements). All 12 unique cross-reference link targets resolve correctly. Four medium-severity findings relate to a missing VitePress sidebar entry and three technical accuracy issues that could mislead teams following the guide. None are blocking, but fixing them would improve the guide's discoverability and reliability. Findings1. Missing VitePress sidebar entry — mediumFile: Per AGENTS.md: "When adding a new doc under The "User Guides" sidebar section (config.ts lines 218–232) uses manual entries, not 2. "Scoped to bug fixes by default" contradicts preceding sentence — mediumFile: The code agent bullet first correctly states the 3. Fix agent "automatic mode" framing is misleading — mediumFile: The guide describes the fix agent as having an "automatic mode" that gets enabled in the Run stage. In reality, the fix agent always auto-triggers on bot-authored PRs when the review agent requests changes. What changes in the Run stage is that the code agent starts producing bot-authored PRs, making auto-fix behavior visible. For human-authored PRs, automatic fixing requires the 4. "Filing issues skill" for retro agent does not exist — mediumFile: The guide says to "pair [the retro agent] with a filing issues skill." The retro agent's harness defines skills 5.
|
|
🤖 Review · |
dbfa7fd to
26f8bb7
Compare
|
🤖 Finished Review · ✅ Success · Started 7:36 AM UTC · Completed 7:46 AM UTC |
|
Finding 1 (VitePress sidebar): Addressed in 1b5d691 — added |
|
🤖 Review · ❌ Terminated · Started 9:36 AM UTC · Ended 9:44 AM UTC |
|
Findings 2-4 addressed in 26f8bb7: Finding 2 (code agent scoping): Updated to "scoped to bugs, docs, and performance issues by default" — reflects all ready-to-code issue types, not just bugs. Finding 3 (fix agent mode): Reframed as "Fix agent on bot-authored PRs" — clarifies auto-trigger behavior on bot PRs vs. label/command requirement on human PRs. Finding 4 (filing issues skill): Changed to "Its post-script files issues from retro findings automatically" — correctly attributes issue filing to the retro agent's post-script, not a separate skill. |
|
🤖 Finished Review · ✅ Success · Started 9:36 AM UTC · Completed 9:44 AM UTC |
1b5d691 to
2409654
Compare
|
🤖 Finished Review · ✅ Success · Started 10:02 AM UTC · Completed 10:15 AM UTC |
|
@maruiz93 could you take a look? |
|
Re: @waynesun09's review noting the triage bot's blocking claim on #4969 — the auto-merge capability has since been implemented: |
|
/fs-fix rebase onto main first (2,122 commits behind, VitePress sidebar will conflict). After rebase, verify filenames — several guides were renamed from customizing-* to configuring-* on main. CODE_AUTO_MERGE env var is in agents repo docs/code.md, ADR 0089 covers risk assessment. |
|
/fs-fix rebase onto main first — all other changes depend on post-rebase file state and link targets |
599c240 to
beefcf4
Compare
|
🤖 Finished Review · ✅ Success · Started 10:25 AM UTC · Completed 10:44 AM UTC Commit: Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $5.72 |
|
Risk Assessment: moderate (2/5) DetailsDocs-only PR from a known contributor adding a new adoption guide (178 lines) plus single-line additions to VitePress config and guides index. No production code, CI, dependency, security, or protected-path changes. Multi-author contention on navigation files slightly elevates git-history tier. |
beefcf4 to
aa1aa3a
Compare
|
🤖 Review · Commit: |
aa1aa3a to
c5b41af
Compare
|
🤖 Review · Commit: |
Site previewPreview: https://e176abe9-site.fullsend-ai.workers.dev Commit: |
Add an incremental adoption guide that walks teams through four stages (Crawl, Walk, Run, Fly) of fullsend usage — from observing default agents to progressive auto-merge and custom agents. Closes fullsend-ai#4969 Signed-off-by: Marta Anon <maruiz@redhat.com> Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> Signed-off-by: Marta Anon <manon@redhat.com>
c5b41af to
1f648a7
Compare
|
🤖 Finished Review · ✅ Success · Started 2:34 PM UTC · Completed 2:53 PM UTC Commit: Runtime: claude · Model: opus → claude-opus-4-6 · Effort: high · Cost: $5.24 |
|
Review skipped — this PR is already merged. The Posted by fullsend post-review check |
Closes #4969
Summary
docs/guides/user/adoption.md— an incremental adoption guide that walks teams through four stages (Crawl, Walk, Run, Fly) of fullsend usage, from observing default agents to progressive auto-merge and custom agentsdocs/guides/README.mdTest plan
🤖 Generated with Claude Code