agentleFS
chakmarebelchakmarebel/federal-proposal-copilot/.claude/skills/submission-summary/SKILL.md

submission-summary

Skill4 starsChanged 3 months agoMITView on GitHub

Use this skill immediately after a solicitation is in hand to produce a one-page summary of exactly what the customer requires us to submit — response format, volume/section structure, page or word limits, pricing artifact, due date, and submission mechanics — before any planning or generation begins.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 2 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run immediately after /new-proposal (or /import-from-capture) and /opportunity-quick-look, and before /proposal-manager.

    line 23
  2. Very specific:

    If inputs/00_priority/ is empty, tell the user to drop the solicitation there first and stop.

    line 50
  3. Very specific:

    Web-form precedence: When submission_mechanism: web-form and inputs/00_priority/portal-format.md exists, the portal format is authoritative …

    line 119
  4. Very specific:

    Flag conflicts: If the solicitation PDF and the portal (or proposal-type.md defaults) disagree on structure or limits, record the conflict …

    line 120
  5. Distinctive:

    Facts only: No win themes, no approach, no solution design.

    line 121
  6. Distinctive:

    One page: If the breakdown runs long, the solicitation's submission instructions are unusually complex — summarize and point to the source …

    line 122
  7. Very specific:

    After writing working/submission-summary.md, stop and emit a decision card (per reference/schemas/decision-card.schema.json) — present it …

    line 126

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Submission Summary Skill

Purpose

State, on one page, exactly what the customer is asking us to submit — before any analysis, planning, or drafting begins.

The shape of the deliverable governs everything downstream: a 3-volume, 10-page effort with a ROM is a different job than a 5-volume full proposal with a priced cost volume. Today that information is scattered — partly in working/proposal-type.md (the generic type), partly buried inside working/proposal-plan.md (extracted late, by proposal-manager). Nobody sees the deliverable shape, plainly and early, before tokens are spent generating against an assumed structure.

This skill produces that artifact: a factual, at-a-glance specification of the submission. It answers, in the first line, "what are we handing over?" — format, volume count, page limit, pricing, due date.

This is the deliverable spec. It is not strategy — no win themes, no approach, no solution.

When to Use

Run immediately after /new-proposal (or /import-from-capture) and /opportunity-quick-look, and before /proposal-manager. It is one of the first things done once the decision to pursue is made.

For web-form submissions, run it after /capture-portal-structure — the captured portal format is the authoritative source for section structure and character limits, and this summary rolls it up into human-readable form.

Recommended workflow:

/new-proposal
/opportunity-quick-look
/capture-portal-structure   (only when submission_mechanism: web-form)
/submission-summary         <<< here — confirm the deliverable shape
/proposal-manager

Inputs

Always read

  1. working/proposal-type.md — the generic type classification, page_target, section_patterns, submission_mechanism, pricing_artifact. This skill records what this specific solicitation requires, which may differ from the generic template defaults.
  2. The solicitation's instructions to offerors, directly — every relevant file in inputs/00_priority/. Specifically the part that tells offerors what to submit: FAR Section L, OTA submission instructions, BAA proposal-preparation instructions, CSO/RWP instructions, SBIR solicitation instructions, or the web-form portal.

Read if relevant

  • inputs/00_priority/portal-format.md — if submission_mechanism: web-form and /capture-portal-structure has run. Authoritative for section structure and character/word limits on web-form submissions.
  • working/quick-look.md — if /opportunity-quick-look ran (its submission-mechanism factor is a starting point).
  • working/section-budgets.md — if /capture-portal-structure produced it.

If inputs/00_priority/ is empty, tell the user to drop the solicitation there first and stop.

Output

Write one file:

working/submission-summary.md

Then surface it to the user for confirmation (see Confirmation Gate).

Structure of working/submission-summary.md

# Submission Summary — [Proposal Name]

**[Headline line — one sentence.]**
Example: "White paper, single document, 5-page limit, ROM pricing, submitted by email — due 2026-06-03 1700 ET."
Example: "FAR full proposal, 3 volumes (Technical 25 pg / Cost no limit / Past Performance 10 pg), priced cost volume, uploaded to SAM.gov — due 2026-07-15 1400 ET."

## Submission Profile

| Field | Value |
|---|---|
| Response format | [white paper / full proposal / solution brief / quad chart / RFI response / ...] |
| Submission mechanism | [email / document-upload / web-form] |
| Number of volumes / files | [N — list them] |
| Due date / time / timezone | [exact] |
| Questions deadline | [date or "none stated"] |
| Submission method / portal | [email address / SAM.gov / portal URL / ...] |
| File format(s) | [PDF / DOCX / ...] |
| File naming convention | [as stated, or "not stated"] |
| Classification / distribution marking | [as required, or "not stated"] |
| POC | [name + email from the solicitation] |

## Volume / Section Breakdown

| Volume / Section | Page or Word Limit | Objective — what it must accomplish / what is evaluated | Required? |
|---|---|---|---|
| [e.g. Volume I — Technical Approach] | [25 pages] | [Demonstrate the technical solution against the SOO] | Required |
| [e.g. Volume II — Cost] | [no limit] | [Priced cost volume with BOEs] | Required |
| [...] | | | |

## Pricing / Cost Deliverable

[What pricing artifact is required — ROM range / OTA milestone schedule / CSO commercial pricing / full FAR cost volume / none — and which volume it lives in. If no pricing is requested, say so explicitly.]

## Required Attachments / Forms

[Resumes, past-performance questionnaires, reps & certifications, OCI disclosures, SF-form X, etc. — or "none stated".]

## Formatting Constraints

[Font, point size, margins, line spacing, page size — only what the solicitation explicitly states. If the solicitation is silent, write "not stated — workspace default applies".]

## What Is NOT Required

[Explicitly call out things sometimes expected but not asked for here — e.g. "No cost volume requested — do not include pricing", "No past performance volume", "No oral presentation". This prevents over-building.]

## Open Questions

[Ambiguities about the submission requirements that should be resolved — ideally before the questions deadline. If none, write "none".]

Rules

  • Extract, do not infer. Quote the solicitation's own words for every limit. If a limit is not stated, write "not stated" — never guess a page count.
  • Page limits are load-bearing. Get them exact. Record whether each limit is per-volume or total, and whether cover pages, tables of contents, glossaries, and appendices count against it. This is the single most consequential field on the page.
  • Web-form precedence. When submission_mechanism: web-form and inputs/00_priority/portal-format.md exists, the portal format is authoritative for section structure and character limits. This summary rolls it up; it does not override it.
  • Flag conflicts. If the solicitation PDF and the portal (or proposal-type.md defaults) disagree on structure or limits, record the conflict in Open Questions — do not silently pick one.
  • Facts only. No win themes, no approach, no solution design. This is a spec of the container, not the content.
  • One page. If the breakdown runs long, the solicitation's submission instructions are unusually complex — summarize and point to the source section, do not transcribe.

Confirmation Gate

After writing working/submission-summary.md, stop and emit a decision card (per reference/schemas/decision-card.schema.json) — present it to the user in chat, and append it as one JSON line to working/decision-cards.jsonl:

{"schema_version":"decision-card.v1","gate_id":"submission-summary-confirmation","proposal_id":"<slug>","generated_by":"submission-summary","generated_at":"<timestamp>","decision":"Confirm the deliverable shape before planning begins.","artifact":"working/submission-summary.md","recommendation":"Confirm as captured","confidence":"<High|Medium|Low>","rationale":"<one or two sentences — note any field that was 'not stated' and had to be left open>","risk_if_wrong":"Every downstream stage is built against the wrong format, volume count, or page budget.","on_approval":"/proposal-manager builds the plan and compliance matrix against this shape","human_time_estimate":"2 minutes","approve_action":"/proposal-manager","review_required":true}

Surface the headline line and the volume/section breakdown with the card and say:

Confirm this is the submission shape before we run /proposal-manager. If the format, volume count, or page limits are wrong, everything downstream is built against the wrong target. A wrong deliverable spec is the most expensive error in the pipeline — cheaper to catch here than at export. Pay particular attention to any field marked "not stated."

review_required is true: confidence is only as good as the solicitation's clarity, so the human confirms the shape — especially the page limits and volume count.

Integration With Downstream Skills

  • /proposal-manager reads working/submission-summary.md as the authoritative source for response requirements. It does not re-extract format, volumes, page limits, or due date — it consumes this file and builds the compliance matrix and proposal structure against the confirmed shape.
  • /proposal-writer produces one draft per volume/section in the breakdown — no more, no fewer.
  • /capture-portal-structure (web-form) and this skill are complementary: the portal capture is the detailed mechanics, this summary is the human-readable rollup.
  • /export-proposal checks the final package against this summary — volume count, file format, naming convention.

Activity Trail

On completion, append one line to working/activity.md:

## <timestamp> — submission-summary — <format>, <N volumes/sections>, <page limit>, <pricing artifact> → working/submission-summary.md

Append one JSON line to working/ai-runs.jsonl per reference/schemas/ai-run.schema.json with job_type: "planning" and skill: "submission-summary".

Final Rule

This skill describes the container, not the content. One page, factual, confirmed by a human before planning begins. That is the whole job.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.