复制安装命令
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
This is the open-source content repository behind
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
来源文件:README.md
This is the open-source content repository behind Skill Store. It stores every approved Agent Skill, the records that go with it, and the automated security audits published with each skill.
This repo is a companion to the Skill Store platform, not the place to submit skills. Skills are added through skillstore.io — its review pipeline writes to this repo automatically. Please do not open a pull request here to add a skill; PRs adding skills will be closed. See Contributing a skill below.
The recommended way to install any skill is the skillstore CLI — one command works for both Claude Code and Codex:
npx skillstore add author/skill-name
For example:
npx skillstore add aiskillstore/code-review
It downloads the skill and drops it into the right skills/ directory for your tool. Claude Code auto-discovers it; for Codex, restart the session.
Prefer to do it by hand, or installing via Claude Web? See the full Installation Guides for every method (CLI, manual, and ZIP upload) and the scope directories (~/.agents/skills/, .claude/skills/, ~/.claude/skills/, .codex/skills/, …).
Submit through the platform — not through a pull request:
SKILL.md.SKILL.md — the skill definition (required, per the Agent Skills spec)LICENSE (recommended)Every submission is scanned automatically before it can be published. The audit flags things like:
eval, exec, raw system commands)Security analysis is report-only: findings inform maintainers and users, but a risk result does not automatically block an otherwise approved skill from being published. See our Security Trust Center for the methodology, limitations, and risk-level definitions.
Live Security Passport example:
.
├── skills/ # Approved, published skills (one folder each, with SKILL.md)
├── pending/ # Submissions awaiting review
├── packages/
│ ├── cli/ # The `skillstore` CLI (npx skillstore add …)
│ └── skillstore/
├── schemas/ # JSON schemas for skill records
├── scripts/ # Maintenance & scoring scripts
└── .github/workflows/ # Submission, audit, and sync automation
The contents of this repo are maintained by Skill Store's automated pipeline. Manual changes are limited to maintainers.
The marketplace catalog is MIT-licensed. Individual skills carry their own licenses — check each skill's LICENSE file.
name: ot-cost-analysis
description: >
Trigger for: Other Transaction or OT should-cost,
cost estimate, cost-share analysis, milestone pricing, funding profile,
proposed-price comparison, prototype price analysis, research OT budget,
production follow-on OT estimate, or OT price-reasonableness support under
10 U.S.C. 4021 or 4022. Build auditable milestone-based .xlsx workbooks using
BLS OEWS wages, CALC+ positioning, GSA Per Diem, materials, cost sharing, and
agreement-specific fees. Use for pre-solicitation budgets and proposal
comparisons. Do NOT use for FAR-based IGCEs, OT project descriptions, grants,
or an Agreements Officer's determination. Requires the bls-oews, gsa-calc,
and, when travel applies, gsa-perdiem MCP servers.Build an auditable, formula-driven cost analysis organized by OT milestone. Separate total project cost, the non-Federal contribution, any consortium or administrative fee, the proposed payment amount, and the Government funding requirement. Use market evidence and arithmetic to support the Agreements Officer without originating the authority path, eligibility finding, significant-participation finding, successful-completion finding, or price-reasonableness determination.
Do not:
When a sentence would conclude the Agreements Officer's judgment, replace it with the source data, the variance, the scenario range, and the decision left open.
Load only the references needed for the active workflow:
/v3/api/ceilingrates/ with keyword= when keyword search is required. Never use q=. Preserve aggregations.labor_category.buckets[*].key and doc_count when discovery is used.detect_latest_year at runtime. Store BLS vintage and agreement start as YYYY-MM; use VALUE(LEFT(...)) and VALUE(MID(...)) for month gap. Never trust a stale constant, use YEAR() on text, or substitute DATEDIF.Select the workflow, then inspect the current session by capability rather than by host-generated tool name.
bls-oews operations detect_latest_year, get_wage_data, metro lookup, and SOC lookup plus gsa-calc operations suggest_contains, exact_search, keyword_search, and igce_benchmark or equivalent schemas.gsa-perdiem operations estimate_travel_cost, lookup_city_perdiem, and get_mie_breakdown only when travel is in scope..xlsx authoring path and Python 3.10 or later with openpyxl for the bundled validators. Prefer a real spreadsheet engine for formula execution.detect_latest_year before wage retrieval. Test Per Diem only when travel first becomes necessary. Apply the three-second keyed-call spacing to tests and production calls.Use when the user supplies a milestone table or the approved handoff from ot-project-description-builder. Preserve milestone IDs, descriptions, deliverables, success criteria, dates, payment type, and user overrides. Ask only for missing cost inputs.
Use when no approved milestone table exists. Derive a candidate milestone structure from the prototype or research objective, TRL progression when relevant, delivery evidence, and schedule. Present the table, ask the user to confirm or amend it, end at that question, and wait. Do not call pricing MCPs before approval.
On the first response, make no tool call. Emit this boundary and stop:
I can build an independent OT should-cost and show the proposed amount's basis, variance, scenario position, and market benchmarks. I cannot originate the Agreements Officer's price-reasonableness determination or negotiation position.
Choose one:
Option A: Comparison data only. I produce neutral tables and a methodology record with no verdict.
Option B: Controlled memo fill. You provide the exact rationale and determination text. I reproduce it verbatim in a DRAFT section and place the neutral evidence beneath it.
Which option?
For Option B, proceed only after the user supplies both the rationale and determination text. Never expand or strengthen it.
Use the host's structured question tool when available. Otherwise use numbered choices and accept numbers, labels, or free text. Batch related missing inputs.
Required before cost-share math:
Required before a build:
Do not guess the authority, statutory path, cost share, location, milestone payment type, fee, proposed-amount basis, or cost-type ceiling.
Treat a table labeled MILESTONE HANDOFF TABLE and identified for OT Cost Analysis as user-reviewed input.
For Workflow A+:
PENDING unless supplied, and funding timing.Read authority-and-cost-sharing.md.
Read data-source-operations.md and labor-and-costing-methods.md.
detect_latest_year, then retrieve metro, state, or national BLS percentiles using the documented fallback ladder.Use the discovery-first operation flow. Keep title-match and experience-match pools separate when useful. Record the pool, count, percentiles, and query date. Sparse or absent results remain a disclosed limitation; they do not authorize an invented specialty premium.
Use user-supplied bills of material, quotes, analogous prices, quantities, and escalation bases. Apply materials escalation from project start to each milestone start using month-based compounding. A production learning curve is used only when the user supplies the method and factor.
Use GSA Per Diem for authorized CONUS travel and the operation rules in the data-source reference. Keep airfare, local transportation, and OCONUS travel on their supplied source bases.
For each milestone:
labor = sum(category hours * category burdened benchmark)
materials = sum(quantity * unit cost * approved escalation)
travel = sum(validated trip costs)
ODCs = sum(other direct cost inputs)
total project cost = labor + materials + travel + ODCs
performer share = total project cost * approved performer ratio
Government share = total project cost - performer share
fee = approved fee base * approved fee rate, or approved fixed amount
Government funding requirement = Government share + Government-paid fee
For cost-type milestones, calculate the same columns at independent should-cost and at the approved ceiling. Label the ceiling view as maximum exposure, not expected actual cost.
Present the per-milestone build and ask for confirmation before final workbook assembly when staffing, materials, or cost allocation was derived rather than supplied.
Vary only assumptions with a documented basis. Keep low, working, and high labor burden, materials, and escalation inputs in workbook cells. Do not ship generic 1.8/2.0/2.2 or materials multipliers as facts unless the user approves them for this analysis.
Read workbook-specification.md in full. Build the seven required sheets and preserve:
Read validation-gates.md.
scripts/validate_workbook.py <workbook> --expected <inputs.json> --engine none.scripts/recompute_expected_values.py <inputs.json>.--engine auto or --engine libreoffice; require the engine layer to reject cached spreadsheet errors anywhere in the recalculated workbook before mapped values are accepted.If no spreadsheet engine is available, state exactly:
Formula structure and independent calculations passed. Formula execution was not independently verified in Excel or LibreOffice.
State which validation layers ran. Summarize the authority and contribution facts as user-supplied inputs, the should-cost range, the Government funding range, key sensitivities, missing evidence, and refresh dates. Do not state a price-reasonableness conclusion unless it is the user's verbatim Option B text, visibly marked DRAFT.
MIT © James Jenrette / 1102tools. Source: github.com/1102tools-dev/federal-contracting-skills
评论 (0)
暂无评论,成为第一个评论者吧!