Audits AI-generated implementation plans for requirements traceability, scope creep, and unverified assumptions.
Works with
Annotates plans inline without rewriting, flagging missing requirement mappings, over-engineering, and risky assumptions with severity levels (critical, warning, info)
Validates technical claims against recent sources via web search when enabled, and uses codebase exploration to verify assumptions
Stops to ask the user for clarification on unresolved assumptions before com
AI-first code editor with Composer
Before installing skills in Cursor, ensure your development environment meets these requirements:
node --versionvalidate-implementation-planExecute the skills CLI command in your project's root directory to begin installation:
Fetches validate-implementation-plan from b-mendoza/agent-skills and configures it for Cursor.
The CLI shows a list of agents. Use arrow keys and space to select Cursor:
Confirm successful installation by checking the skill directory location:
Restart Cursor to activate validate-implementation-plan. Access via /validate-implementation-plan in your agent's command palette.
We perform automated surface-level scans (Gen AI Scanner, Socket, Snyk) during installation. These checks detect common vulnerabilities but do not guarantee complete security. Always review skill source code and verify the publisher's reputation before production use.
Skills execute code in your environment. Always review source, verify the publisher, and test in isolation before production.
Submit your Claude Code skill and start earning
Create detailed user stories, acceptance criteria, and feature specs
Example
Generate user stories for 'password reset feature' with acceptance criteria, edge cases, and test scenarios
Reduce spec writing time by 50%, ensure comprehensive coverage
Research competitors, compare features, identify gaps
Example
Analyze 5 competitor products, create feature comparison matrix, suggest differentiation opportunities
Complete competitive research in 2 hours instead of 2 days
Evaluate features using frameworks (RICE, ICE, Kano) and create prioritized backlogs
Example
Score 20 feature ideas using RICE framework, generate prioritized roadmap with rationale
0
total installs
0
this week
0
upvotes
Run in your terminal
0
installs
0
this week
—
stars
You are the orchestrating agent for an implementation plan audit. You coordinate a team of specialist subagents — you never perform the audit work yourself. Your context window is precious: dispatch, collect concise results, and synthesize.
| Position | Name | Type | Default | Description |
|---|---|---|---|---|
$0 |
plan-path |
string | (required) | Path to the plan file to audit |
$1 |
write-to-file |
true / false |
true |
Write the annotated plan back to the file at $0. Set to false to print to conversation only. |
$2 |
fetch-recent |
true / false |
true |
Use WebSearch to validate technical assumptions against recent sources (no older than 3 months) |
| Subagent | Path | Purpose |
|---|---|---|
technical-researcher |
./subagents/technical-researcher.md |
Validates technical claims in the plan against current web sources |
requirements-extractor |
./subagents/requirements-extractor.md |
Extracts numbered source requirements from the user's original request and related context |
requirements-auditor |
./subagents/requirements-auditor.md |
Audits every plan section for traceability back to source requirements |
yagni-auditor |
./subagents/yagni-auditor.md |
Audits every plan section for scope creep, over-engineering, and premature abstraction |
assumptions-auditor |
./subagents/assumptions-auditor.md |
Identifies and attempts to verify assumptions; returns unresolved items for orchestrator to clarify |
plan-annotator |
./subagents/plan-annotator.md |
Merges all annotations into the original plan and compiles the audit summary |
Execute these steps in order. Pass structured data between steps — never rely on ambient context.
These subagents are co-located in this skill's subagents/ directory —
they are not auto-discovered from .claude/agents/. To dispatch one:
Read the subagent's .md file from the path in the registry above.Task tool, passing the subagent's file content as the
system prompt and your task-specific instructions (inputs, expected
output format) as the prompt.AskUserQuestion is not available inside subagents. This is a Claude Code platform limitation — the tool silently fails when called from a Task-spawned subagent. That is why the assumptions-auditor escalates unresolved items back to the orchestrator (Step 5), where AskUserQuestion works normally. Do not attempt to move user interaction into any subagent.
plan_text = Read($0)
Store plan_text as the canonical input. Every subagent receives this
verbatim — never paraphrase or summarize the plan.
Skip this step entirely when $2 is false.
Dispatch technical-researcher (read ./subagents/technical-researcher.md,
pass via Task tool) with:
plan_text — the full plan contentCollect: research_findings — a structured list of validated/invalidated
claims with source URLs and dates. This is passed to downstream auditors
as supplementary evidence.
Dispatch requirements-extractor with:
plan_textCollect: requirements_list — a numbered list of requirements and
constraints. This is the reference baseline for all auditors.
Dispatch each auditor sequentially. Every auditor receives:
plan_textrequirements_listresearch_findings (empty string if Step 2 was skipped)Dispatch requirements-auditor. Collect: req_annotations — a list of
annotations with section references, severity levels, and requirement
citations.
Dispatch yagni-auditor. Collect: yagni_annotations.
Dispatch assumptions-auditor. Collect two things:
assumption_annotations — annotations for assumptions that were
resolved through plan text, codebase search, or web researchunresolved_assumptions — a list of assumptions that could not be
verified, each with:
section: which plan section it appears inassumption: what is being assumedquestion: a proposed question to ask the userdraft_annotation: the annotation to use if the assumption is
confirmed as riskyFor each item in unresolved_assumptions, use AskUserQuestion to ask
the user the proposed question. Record the user's answer.
After collecting all answers, re-dispatch assumptions-auditor with:
unresolved_assumptions listCollect: resolved_annotations — final annotations for the previously
unresolved items, with severity adjusted based on user input.
Merge resolved_annotations into assumption_annotations.
Dispatch plan-annotator with:
plan_textrequirements_listreq_annotationsyagni_annotationsassumption_annotations (now includes resolved items)Collect: annotated_plan — the complete output document.
$1 is true or omitted: write annotated_plan to the file at $0
using Write.$1 is false: output annotated_plan to the conversation.AskUserQuestion gets no response or an ambiguous answer, record
the assumption as an Open Question in the summary — do not guess.$0 cannot be read, stop immediately and inform
the user.The final output follows this structure (produced by plan-annotator):
## Source Requirements
1. <requirement>
2. <constraint>
...
---
## Annotated Plan
<original plan content reproduced exactly>
// annotation made by <Expert Name>: <severity> <text referencing requirement number>
...
---
## Audit Summary
| Category | 🔴 Critical | 🟡 Warning | ℹ️ Info |
| ------------------------- | ----------- | ---------- | ------- |
| Requirements Traceability | N | N | N |
| YAGNI Compliance | N | N | N |
| Assumption Audit | N | N | N |
**Confidence**: ...
**Resolved Assumptions**:
- <assumption> — User confirmed: <answer>. Annotation adjusted to <severity>.
**Open Questions**:
- <only items where the user chose not to answer or the answer was ambiguous>
Make data-driven prioritization decisions faster
Draft PRDs, status updates, and stakeholder presentations
Example
Create executive summary of Q3 roadmap, monthly progress report, feature launch announcement
Save 3-5 hours/week on communication overhead
Prerequisites
Time Estimate
30-60 minutes to see productivity improvements
Steps
Common Pitfalls
✓ Do
✗ Don't
💡 Pro Tips
✓ Use when
Use for user story writing, competitive research, roadmap prioritization, stakeholder communication, and PRD drafting. Best for reducing repetitive documentation and research work.
✗ Avoid when
Avoid for strategic product vision (requires deep customer empathy), pricing decisions (needs market and financial expertise), or when face-to-face customer discovery is more valuable than speed.
mattpocock/skills
mattpocock/skills
parcadei/continuous-claude-v3
cursor/plugins
ailabs-393/ai-labs-claude-skills
pproenca/dot-skills
validate-implementation-plan has been reliable in day-to-day use. Documentation quality is above average for community skills.
Registry listing for validate-implementation-plan matched our evaluation — installs cleanly and behaves as described in the markdown.
I recommend validate-implementation-plan for anyone iterating fast on agent tooling; clear intent and a small, reviewable surface area.
Useful defaults in validate-implementation-plan — fewer surprises than typical one-off scripts, and it plays nicely with `npx skills` flows.
Solid pick for teams standardizing on skills: validate-implementation-plan is focused, and the summary matches what you get after install.
validate-implementation-plan fits our agent workflows well — practical, well scoped, and easy to wire into existing repos.
Keeps context tight: validate-implementation-plan is the kind of skill you can hand to a new teammate without a long onboarding doc.
validate-implementation-plan is among the better-maintained entries we tried; worth keeping pinned for repeat workflows.
Keeps context tight: validate-implementation-plan is the kind of skill you can hand to a new teammate without a long onboarding doc.
We added validate-implementation-plan from the explainx registry; install was straightforward and the SKILL.md answered most questions upfront.
showing 1-10 of 41