Works with the AI Agents you already use
attn
Audit technical text, specs, and docs for logical conflicts, ambiguity, gaps, and structural bloat.
Free
attn
Example session with this skill installed
Audit this deployment process
- Pull latest image.
- Stop container.
- Ensure the .env is updated before step 2.
- Restart. Use the attn skill.
- Read your context and instructions
- Compiled the attn
I have audited the deployment process using the requested lenses.
Audit Results
Conflicts — clean
Ambiguity — clean
Gaps — clean
Duplicates — clean
Priorities — clean
Order — 3. Ensure the .env is updated before step 2.
The instruction for the .env update is placed after the step it modifies. The user may stop the container (Step 2) before realizing they needed to check the environment configuration first.
Noise — clean
Bloat — clean
Verdict: revise
Next steps
- Move the .env update requirement to Step 1.
- Re-index the sequence to ensure dependencies are linear.
Connects securely to your tools. The creator never sees your data.
About this skill
Communication debt ruins technical execution. This skill audits any text—technical specs, documentation, or internal plans—to identify friction points that lead to misinterpretation or wasted cycles. It acts as a cold-eyed reviewer for your written output, catching logical traps and structural rot before they reach your team.
What it does
- Conflict Detection — Identifies contradictory rules, inconsistent definitions, and steps that cannot coexist.
- Ambiguity Triage — Flags vague terms, fuzzy thresholds, and undefined scopes where the reader is forced to guess.
- Structural Gaps — Pinpoints missing handlers for enumerated cases or references to non-existent sections.
- Redundancy & Noise Removal — Surfaces duplicates, motivational fluff, and boilerplate that hides the signal.
- Flow Optimization — Analyzes sequence to ensure terms are defined before use and stop conditions precede actions.
How it works
- Provide source — Paste the text, documentation, or technical plan you need audited.
- Lens Analysis — The agent applies specific filters including Conflicts, Ambiguity, Gaps, and Order.
- Evidence Mapping — Problems are mapped to specific quotes or line numbers with a clear explanation of the friction.
- Final Verdict — Receives a status of clean, minor, or revise based on the severity of the findings.
Frameworks & tools
This skill is framework-agnostic and works with Markdown, plain text, technical specifications, and process documentation.
Why this beats prompting it yourself
Standard LLM feedback is often overly polite or generic. This skill uses a specialized lens-based heuristic to force the agent to find specific logical failures that generic "please review this" prompts overlook.
Use cases
- Auditing internal technical specs for logical contradictions before development starts.
- Reviewing onboarding documentation to identify missing steps or circular references.
- Pruning verbose architecture proposals to remove noise and highlight critical priorities.
- Checking PR descriptions or READMEs for sequence errors and undefined terms.
Known limitations
This skill focuses on internal logic and clarity based on the text provided; it does not perform external fact-checking or verify domain-specific completeness outside the source.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Technical Documentation.
Evaluates text based only on provided source. Does not verify external technical accuracy or perform third-party fact-checking. Best for logic and clarity.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 27 days ago
- Free with an Agensi account
Permissions
No special permissions declared or detected