agentsclimarketplace

Cvg plan review feedback

Skill gomilesf/convergo/skills-src/cvg-plan-review-feedback

Disciplined software engineering for coding agents: plan → review → build loops that actually terminate, with bounded rounds, fresh-reviewer exit gates, and TDD. Plugin for Claude Code and Codex.

Install
npx -y skills add gomilesf/convergo --skill cvg-plan-review-feedback

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 7 stars7 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Handle plan-review blocker findings returned to the planner session. Validate feedback, revise plan-owned artifacts, or stop for missing decisions.

SKILL.md

3.3 KB, 645 tokens by cl100k_base, as published. Nobody here has run it

Plan Review Feedback

Use this skill in the planner session after a plan reviewer returns blocker findings.

The input is the reviewer blocker findings appended to the prompt, plus the original planning goal, source prompt, current plan, behavior contract if present, and relevant codebase evidence.

Do not treat reviewer findings as an edit list. First decide whether each finding is valid under the accepted goal and current code reality, then revise only plan-owned artifacts.

If the task context provides no coordination channel (standalone use), the coordination channel is the user: stop and ask directly.

Process

1. Reload planning authority

Read:

  • original user goal, issue, acceptance criteria, and non-goals,
  • current plan document,
  • behavior contract or linked design notes, if present,
  • relevant codebase evidence for each reviewer finding,
  • exact reviewer blocker findings.

2. Evaluate each blocker

For each finding, decide whether it is:

  • Valid plan gap: the plan misses a required surface, lifecycle, invariant, matrix cell, ownership boundary, or verification gate.
  • Source-backed contract gap: the source authority defines the behavior, but the plan or contract failed to capture it.
  • Decision-required contract gap: neither the plan nor source authority defines the behavior clearly enough to proceed.
  • Reviewer clarification needed: the finding cannot be mapped to concrete plan or code evidence after re-reading the relevant context.
  • Invalid or out of scope: the finding is stale, already covered, contradicts the accepted goal, or asks for work outside the agreed scope.

3. Revise only planner-owned artifacts

For valid plan gaps and source-backed contract gaps:

  • update the plan, contract, surfaces, invariant matrix, slices, or verification gates as needed,
  • keep the revision scoped to the accepted goal,
  • preserve reviewer finding ids or exact wording so re-review can check them.

For decision-required contract gaps:

  • stop and callback through the coordination channel,
  • state the missing decision and the blocked plan section,
  • do not invent behavior.

For reviewer clarification:

  • callback with the exact ambiguity and evidence checked.

For invalid or out-of-scope findings:

  • keep the plan unchanged for that finding,
  • explain why it does not apply.

Output

Before editing, write a compact intake summary:

Plan review feedback intake:
1. Finding: <reviewer finding id or short quote>
   Decision: <valid plan gap / source-backed contract gap / decision-required contract gap / clarification needed / invalid or out of scope>
   Action: <plan edit / contract edit / stop for decision / ask reviewer / no change>
   Evidence: <plan/code reference>

After acting, report:

  • findings addressed,
  • plan or contract sections changed,
  • findings rejected or needing clarification,
  • decisions still blocking the planner,
  • verification gates added or changed.

Rules

  • Do not edit implementation code.
  • Do not hide implementation bugs by reducing plan scope.
  • Do not invent behavior for contract gaps.
  • Do not silently drop invalid or out-of-scope findings.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most plan spec skills give in 645 tokens

Counted across 1,099 of the 1,860 authors here whose files we hold, read 2026-08-07

  • Ask one question at a timein 51 of 1099
  • Break plans into vertical slicesin 29 of 1099, across 11 files
  • Publish issues in dependency orderin 27 of 1099, across 9 files
  • Iterate until user approves the breakdownin 25 of 1099, across 7 files
  • Explore the repository to understand the codebase statein 24 of 1099, across 7 files
  • Use domain glossary vocabularyin 23 of 1099, across 5 files
  • Apply correct triage labels to published issuesin 23 of 1099, across 5 files
  • Prefer AFK slices over HITLin 22 of 1099, across 7 files
  • Write a specification before writing any codein 22 of 1099, across 14 files
  • Write failing tests before implementation codein 22 of 1099, across 20 files
  • Ask clarifying questions until requirements are concretein 21 of 1099, across 13 files
  • Respect existing architecture decision recordsin 20 of 1099, across 5 files

Said here and by no other author read

  • Validate each blocker finding under the accepted goal
  • Revise only plan-owned artifacts
  • Stop for decisions through the coordination channel
  • Explain rejected findings
  • Write a compact intake summary before editing
  • Report addressed findings after acting

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.

Keep looking

Skills are one crate of 326,970. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.