# Reviewing Color Teams > Runs proposal color team reviews for a US Federal bid — Blue, Pink, Green, Red, Gold, and White Glove. Use when the user needs a color team review; a Blue, Pink, Green, Red, Gold, or White Glove (book check) review; a proposal or cost-volume review; an evaluator-perspective assessment; a mock evaluation; or a proposal scored before submission. The Red Team scores the proposal as a source-selectio… - **Type:** Skill - **Install:** `agentstack add skill-danielkinneyspears-federal-proposal-skills-reviewing-color-teams` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [danielkinneyspears](https://agentstack.voostack.com/s/danielkinneyspears) - **Installs:** 0 - **Category:** [Communication](https://agentstack.voostack.com/c/communication) - **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/reviewing-color-teams ## Install ```sh agentstack add skill-danielkinneyspears-federal-proposal-skills-reviewing-color-teams ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # Reviewing Color Teams Run a proposal review the way a federal source-selection evaluator would. Color team reviews catch problems while they are still cheap to fix. Their value depends entirely on one thing: producing findings that are scored, prioritized, and specific enough to act on. A review that returns vague impressions wastes the most expensive afternoon in the proposal schedule. This skill runs the review against an explicit standard and produces a findings matrix the team can work in recovery. ## When to use this skill Use this skill to run any of the proposal color team reviews: - Blue Team: early, on the solution and win strategy, before or during solutioning. - Pink Team: on storyboards and an early draft (~30–50%). - Green Team: on the cost/price volume and the pricing strategy. - Red Team: near-final (~90%), scoring as the government evaluator would. - Gold Team: final, executive go/no-go before submission. - White Glove: final production / book check of the assembled package. Determine which review fits the work and its maturity (the skill's intake does this). This skill reviews; it does not fix. The fixes are recovery work, run by `running-review-recovery` with the drafting skills. ## Inputs **Required:** - The proposal content to review — storyboards (`11-proposal-outline.md`) for a Pink Team; drafted sections (`12-sections/` and the volume artifacts) for Red and Gold. - `10-compliance-matrix.md`: the requirements the proposal is checked against. - The solicitation Sections L and M — the evaluation standard. **Preferred upstream artifacts:** - `05-win-strategy.md`: to assess whether themes and discriminators landed. **Without Section M**, a Red Team cannot score properly — Section M *is* the scoresheet. Obtain it before running a Red Team. Read `../../shared/glossary.md`, `../../shared/proposal-lifecycle.md`, and `../../shared/pursuit-workspace.md` if not already read this session. ## Intake Ask these as a numbered list. 1. **Which review.** Which color team — Blue, Pink, Green, Red, Gold, or White Glove? If unsure, describe what is being reviewed and its maturity: solution/win strategy → Blue; storyboards and early draft → Pink; the cost volume → Green; a near-final full draft → Red; the final proposal → Gold; the assembled submission package → White Glove. 2. **Scope.** The whole proposal or specific volumes/sections? Which artifacts are available to review? 3. **The standard.** Are Section L, Section M, and the compliance matrix available? For a Red Team, what are the Section M factors, subfactors, their relative importance, and the evaluation approach (tradeoff or LPTA)? 4. **Win strategy.** Is `05-win-strategy.md` available, so the review can check whether the themes and discriminators are present and persuasive? 5. **Recovery time.** How much time remains before the next milestone or submission? (This shapes how findings are prioritized.) ## Workflow ``` Color Team Review: - [ ] Step 1: Confirm the review type and load the standard - [ ] Step 2: Run the review for that team's lens - [ ] Step 3: Record each finding with location, severity, and a fix - [ ] Step 4: Score (Red Team: rate against Section M) - [ ] Step 5: Prioritize findings for recovery - [ ] Step 6: Write the review (20-color-review-.md) ``` **Step 1: Review type and standard.** Confirm which review this is. Load the applicable standard from `references/color-team-standards.md` (the lens and maturity expectation for that team) and, for a Red Team, `references/evaluator-perspective.md` (how to score as a source-selection evaluator). **Step 2: Run the review.** Review every in-scope piece of the work through that team's lens: - Blue Team: solution and win strategy. Is the solution sound and does it deliver the discriminators? Are the win themes real and substantiated? This review happens before heavy drafting, while the approach can still change. - Pink Team: strategy and structure. Do the storyboards and early draft follow Section L? Is each section's theme present and tied to a hot button and a Section M factor? Is the compliance approach sound? Is the proposal on the right track *before* heavy drafting? - Green Team: cost and price. Is the pricing strategy sound? Is the basis of estimate sufficient? Is the cost realistic and consistent with the technical and management volumes? Is the cost volume compliant with Section L? - Red Team: the evaluator's view. Read each section as the government evaluator who scores it. Is every requirement compliant? Where are the strengths, weaknesses, significant weaknesses, and deficiencies? Would this win the tradeoff? - Gold Team: executive readiness. Is it compelling, coherent across volumes, compliant, and ready to submit? Go or no-go? - White Glove: production / book check. Is the assembled package complete, consistently formatted, correctly produced, and free of production defects? (Substantive conformance is `checking-submission-compliance`'s job; the White Glove review is the human quality pass over the produced package.) **Step 3: Record findings.** Every finding is recorded with: its exact location (volume, section, requirement ID); what is wrong; its severity; and a specific recommended fix. A finding without a location and a fix is an opinion, not a finding — do not record it that way. **Step 4: Score.** For a Red Team, assign ratings as an evaluator would, using `evaluator-perspective.md`: rate each Section M factor (adjectival rating and the strengths/weaknesses behind it), assess past performance confidence, and give an integrated read of whether the proposal wins. Reviews other than the Red Team give a readiness assessment rather than a formal score, but still state plainly whether the work is on track. **Step 5: Prioritize.** Sort findings for recovery: deficiencies and significant weaknesses first (a deficiency makes the proposal unawardable until fixed), then weaknesses, then strengths to amplify, then minor issues. Given the recovery time available, mark which findings *must* be fixed before the next milestone. **Step 6: Write.** Write `20-color-review-.md` (e.g., `20-color-review-red.md`) using `templates/color-review.md`. ## Output One artifact per review: `20-color-review-.md` in the pursuit workspace. Present the user with: the overall assessment (for a Red Team, the scored verdict), the deficiencies and significant weaknesses that must be fixed, the prioritized recovery list, and (honestly) whether the proposal is on track. ## References - `references/color-team-standards.md`: the six reviews, their maturity expectations and lenses, the findings format, and the recovery handoff. - `references/evaluator-perspective.md`: scoring as a source-selection evaluator: the definitions of strength, weakness, significant weakness, and deficiency; adjectival ratings; past performance confidence; and scoring against Section M. - `templates/color-review.md`: the color review artifact template and findings matrix. ## Guardrails - **Findings must be actionable.** Every finding has a location, a severity, and a specific fix. Vague comments ("strengthen this section") are not findings. - **Score against Section M, not taste.** A Red Team rates the proposal against the solicitation's stated evaluation criteria — not against what the reviewer personally would prefer. - **Be the adversary.** A Red Team that is gentle is worthless. Find the weaknesses and deficiencies now, while the team can still fix them, because the government's evaluator will not be gentle. - **Do not rewrite.** This skill reviews and prescribes fixes. The fixes are recovery work for the drafting skills. A reviewer who rewrites stops reviewing. - **Honest verdict.** If the proposal is not on track, say so plainly. A review that softens a bad verdict denies the team the chance to recover. ## 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-reviewing-color-teams - 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%.