# Analyzing Debriefs > Prepares for and analyzes a US Federal procurement debrief. Use when the user needs to request a debrief, prepare debrief questions, analyze a government debrief, conduct win/loss analysis, or capture lessons learned from a federal bid. Structures the debrief request, interprets what the government said about the evaluation, and turns it into lessons that feed future bid/no-bid and capture decisi… - **Type:** Skill - **Install:** `agentstack add skill-danielkinneyspears-federal-proposal-skills-analyzing-debriefs` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [danielkinneyspears](https://agentstack.voostack.com/s/danielkinneyspears) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** Apache-2.0 - **Upstream author:** [danielkinneyspears](https://github.com/danielkinneyspears) - **Source:** https://github.com/danielkinneyspears/federal-proposal-skills/tree/main/skills/analyzing-debriefs ## Install ```sh agentstack add skill-danielkinneyspears-federal-proposal-skills-analyzing-debriefs ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Analyzing Debriefs Close the loop on a pursuit. After award (win or lose) the government's debrief is the single best source of truth about how the proposal was actually evaluated. Most teams underuse it: they request it late, ask weak questions, or read a loss debrief through sour grapes. This skill helps request the debrief properly, interpret what the government said, and convert it into lessons that make the next bid better. ## When to use this skill Use this skill after a federal award decision, in two modes: - Before the debrief: to request it on time and prepare strong questions. - After the debrief: to analyze what the government said and capture win/loss lessons learned. Use it for wins as well as losses. A win debrief reveals what worked and where the proposal was weaker than its score suggested — both valuable. ## Inputs **Before the debrief:** - The award notification and the solicitation (for the debrief rules). - The pursuit workspace artifacts — `01-bid-decision.md`, `05-win-strategy.md`, `20-color-review-*.md` — so questions target what the team most needs to learn. **After the debrief:** - The government's debrief content (the written debrief, slides, or notes from an oral debrief). - The same pursuit workspace artifacts, to compare what the team believed against what the government found. Read `../../shared/glossary.md`, `../../shared/federal-solicitation-primer.md`, and `../../shared/pursuit-workspace.md` if not already read this session. ## Intake Ask these as a numbered list. 1. **Mode.** Is this preparing to request a debrief, or analyzing a debrief already received? 2. **Outcome.** Won or lost? If lost, is the awardee known? 3. **Timing.** When was the award notification received? (Debrief request windows are short and strict — see the method file.) 4. **Debrief type.** Is the debrief written, oral, or a combination? What did the government provide or schedule? 5. **What the team wants to learn.** What are the open questions — why the proposal scored as it did, how it compared, what the deciding factors were? ## Workflow: requesting a debrief ``` Debrief Request: - [ ] Step 1: Confirm the debrief rules and the request deadline - [ ] Step 2: Draft a timely debrief request - [ ] Step 3: Prepare focused debrief questions ``` **Step 1: Rules and deadline.** From the solicitation and the award notice, confirm the debrief rules and (critically) the deadline to request a debrief. The window is short. Use `references/debrief-method.md`. A debrief requested late may be denied, and a timely debrief can affect other downstream timelines. **Step 2: Request.** Help draft a prompt, professional written request within the deadline. **Step 3: Questions.** Prepare focused questions: how the proposal was evaluated against each Section M factor, the strengths and weaknesses identified, and (within what the government may disclose) how it compared and the basis for the award decision. Good questions are specific and tied to the evaluation factors; vague questions get vague answers. ## Workflow: analyzing a debrief ``` Debrief Analysis: - [ ] Step 1: Capture what the government said, factor by factor - [ ] Step 2: Compare government findings to the team's own view - [ ] Step 3: Separate observations from possible protest issues - [ ] Step 4: Extract win/loss lessons learned - [ ] Step 5: Write the analysis (30-debrief-analysis.md) ``` **Step 1: Capture.** Record what the government said, organized by Section M factor: the strengths, weaknesses, significant weaknesses, and deficiencies it identified, its ratings, and (to the extent disclosed) the comparison and the award rationale. Capture it factually, in the government's words, before interpreting. **Step 2: Compare.** Set the government's findings against the team's own view — the win strategy, the Red Team's findings (`20-color-review-*.md`), the bid/no-bid rationale. Where did the government agree with the Red Team? Where did it find something the reviews missed? Where did a believed discriminator fail to score? The gaps between what the team thought and what the government found are the richest lessons. **Step 3: The protest boundary.** A debrief sometimes surfaces something that looks like a procurement irregularity. This skill **does not provide legal advice and does not assess protest merit.** If the debrief suggests the evaluation may not have followed the stated criteria or the solicitation, the skill flags it plainly as a matter for contracts counsel and notes that protest deadlines are short — it does not advise whether or how to protest. That decision belongs to counsel. **Step 4: Lessons learned.** Extract concrete, reusable lessons. For each, name what to do differently and which future activity it changes — a bid/no-bid criterion, a capture action, a proposal practice, a pricing posture. Avoid sour grapes: a loss attributed entirely to bad luck or an unfair government teaches nothing. Be as honest about a win's weak spots as about a loss's. **Step 5: Write.** Write `30-debrief-analysis.md` using `templates/debrief-analysis.md`. ## Output One artifact: `30-debrief-analysis.md` in the pursuit workspace — the captured debrief, the comparison, any flagged counsel item, and the lessons learned. Present the user with: the headline reasons for the outcome, the most important gaps between the team's view and the government's, any item flagged for counsel, and the top lessons for the next pursuit. ## References - `references/debrief-method.md`: debrief rights and timing, preparing questions, reading a debrief, win/loss lessons, and the protest boundary. - `templates/debrief-analysis.md`: the debrief analysis artifact template. ## Guardrails - **Mind the deadline.** Debrief request windows are short and strict. When helping request a debrief, lead with the deadline. - **No legal advice.** This skill does not assess protest merit or advise whether to protest. It flags possible irregularities for contracts counsel and notes that protest timelines are short. Counsel decides. - **Capture before interpreting.** Record the government's actual words first; analyze second. Do not reshape what the government said into what the team wishes it had said. - **No sour grapes.** A loss explained away as bad luck produces no lesson. Find the controllable factors. - **Lessons must be actionable.** Every lesson names what changes and where — otherwise it is a feeling, not a lesson. ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [danielkinneyspears](https://github.com/danielkinneyspears) - **Source:** [danielkinneyspears/federal-proposal-skills](https://github.com/danielkinneyspears/federal-proposal-skills) - **License:** Apache-2.0 Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-danielkinneyspears-federal-proposal-skills-analyzing-debriefs - Seller: https://agentstack.voostack.com/s/danielkinneyspears - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.