- Home
- Skills
- Finance & Accounting
- Draft Repair
Works with the AI Agents you already use
Draft Repair
A systematic procedure for reconciling, outlining, and repairing disjointed proposal drafts against SOW requirements.
$5
Draft Repair
Example session with this skill installed
Repair section_id 4.2 using sow.md and messy_draft.docx. Page budget is 3. Weave 'Zero Trust' and 'Scale' themes. Proof points: 99.9% uptime, 50k users. Output to repaired_4.2.md.
- Read your context and instructions
- Compiled the draft repair
I have repaired the draft for Section 4.2. I performed a full inventory of the 12 blocks in the source draft, identified three stray duties that were removed, and mapped the remaining content to PWS 4.2.1 through 4.2.4.
Section 4.2: Technical Infrastructure
Budget: 3 pages | Estimated: 2.1 pages | Word Count: 945 (was 1,250)
4.2.1 System Availability (PWS 4.2.1)
The [Offeror] ensures 99.9% system uptime using a load-balanced architecture across three availability zones. This mechanism provides a self-healing infrastructure that scales automatically to support up to 50,000 concurrent users (PWS 4.2.1).
4.2.2 Security Architecture (PWS 4.2.2)
The [Offeror] implements a Zero Trust architecture for all internal service communication. Every request is authenticated and authorized via mTLS, ensuring that identity is verified regardless of network location (PWS 4.2.2).
... [Additional sections following SOW order]
Compliance Table
| SOW ID | Modal | Subsection | Evidence in this section | Gap |
|---|---|---|---|---|
| 4.2.1 | shall | 4.2.1 System Availability | Uptime proof + scaling mechanism | None |
| 4.2.2 | shall | 4.2.2 Security Architecture | Zero Trust implementation | None |
| 4.2.3 | should | 4.2.3 Legacy Support | Compatibility statement | None |
Repair Log
- Blocks Dropped: 3 (2 slogans regarding "market leadership", 1 stray duty regarding janitorial services).
- Modality Restored: 4.2.3 was promoted to 'shall' in the draft; restored to 'should'.
- Open Items: None.
Next steps
- Review the scaled user count against the updated hardware list.
- Verify the mTLS implementation matches the current internal security policy.
- Export the markdown to the final proposal template.
Connects securely to your tools. The creator never sees your data.
About this skill
The problem
Proposal drafts often become disjointed through multiple contributors, repeating content while missing mandatory requirements. Manual reconciliation against a Statement of Work (SOW) is slow and prone to compliance gaps.
What it does
- Inventories existing draft blocks to flag slogans, repetitions, and invented metrics.
- Reconciles draft content against specific SOW identifiers to find missing "shalls."
- Generates a rigid outline based on SOW order before writing any prose.
- Rewrites prose to prioritize evaluator readability with inline SOW tagging.
- Calculates page-budget estimates to ensure the repair fits solicitation constraints.
Why this beats prompting it yourself
Standard prompts often try to "fix everything at once," leading to hallucinations or lost requirements. This skill enforces a mandatory six-step procedure—inventory, reconciliation, and outlining—before a single sentence of prose is written, ensuring 100% compliance coverage.
Use cases
- Repairing a "Pink Team" draft that drifted from the SOW structure.
- Auditing a technical section to ensure every "shall" has a tagged mechanism.
- Stripping marketing fluff and slogans from engineer-written content.
- Weaving specific win themes into a compliant narrative without adding bulk.
Known limitations
Does not generate first drafts from a blank page. Requires a pre-existing draft and specific SOW items to function.
Requires an existing draft and SOW extract. Does not handle book-level compliance matrices or past-performance card creation. Scope is limited to repairing one section at a time.
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 any payment.
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
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 2 months ago