Agent Claude Code
Subagent — tests skills against realistic scenarios and evaluates output against pre-defined criteria; returns a structured pass/fail report.
Agent Claude Code
Subagent — tests skills against realistic scenarios and evaluates output against pre-defined criteria; returns a structured pass/fail report.
Agent Claude Code
Subagent — runs the final quality gate and publishes a tested skill to Xeebs/grimoire on GitHub with a Discussion announcement.
Agent Claude Code
Subagent — scans web sources for AI skill gap signals in Finance, Legal, and Consulting; also monitors GitHub for feedback on published skills.
Agent Claude Code
Subagent — designs complete, novel AI skills from researched signals; produces SKILL.md, README.md, and test scenarios.
Command Claude Code
Publish all skills in skills/ that have passed testing and are marked ready. For each skill.
Command Claude Code
Address feedback from a GitHub Issue on a published Grimoire skill. Usage: /project:fix-issue {issue-number}.
Command Claude Code
Trigger a Grimoire research cycle. Scan GitHub, Hacker News, Reddit, and practitioner blogs for signals on AI workflow gaps in Finance, Legal, and Consulting. Score each signal (High/Medium/Low) and append findings to pipeline/queue.md. Skip any signals already present in the queue. Report a summary of new signals…
Command Claude Code
Review a staged skill against the Grimoire quality bar before publishing. Usage: /project:review {skill-path}.
Command Claude Code
Show the current Grimoire pipeline status. Read pipeline/state.json and pipeline/queue.md. Report.
Settings file Claude Code
Agent settings declaring 12 allowed tools.
Skill Claude CodeCodex
Spawns the publisher subagent to run the final quality gate and publish a tested skill to Xeebs/grimoire on GitHub.
Skill Claude CodeCodex
Spawns the skill-designer subagent to design a complete novel skill from the top-priority signal in the pipeline queue.
Skill Claude CodeCodex
Spawns the researcher subagent to scan web sources for AI skill gap signals in Finance, Legal, and Consulting, and to monitor GitHub feedback on published skills.
Instructions file
Claude Code instructions for Xeebs/grimoire, covering grimoire — ai skills lab, mission, subagent architecture, autonomous heartbeat protocol and phase 1 — resume check.
Skill Claude CodeCodex
For strategy consultants building competitive benchmarking slides, normalizes competitor metrics across sources with definition reconciliation, identifies the 2–3 strategically significant gaps tied to the client question, and outputs a slide-ready brief with a narrative headline, comparison table, and callout boxes.
Skill Claude CodeCodex
For strategy consultants and business analysts: given a target market description, industry statistics inputs from named benchmark sources, and a client ICP profile, constructs a bottoms-up sizing model and a parallel top-down model, calculates TAM/SAM/SOM at each level, identifies and explains the reconciliation gap…
Skill Claude CodeCodex
For PMO consultants and integration managers: given a set of free-text workstream status updates from post-merger integration track leads, normalizes each to a consistent RAG status using defined criteria, detects optimism bias where self-reported GREEN conflicts with described blockers, identifies cross-workstream…
Skill Claude CodeCodex
Synthesizes 10–30 stakeholder interview notes into a MECE issue tree, detects cross-group contradictions, and produces a hypothesis-driven recommendation brief mapped to the client's strategic question — for management consultants transitioning from field research to synthesis workshop.
Skill Claude CodeCodex
Extracts named financial metrics from CIM PDFs into a model-ready table with data-provenance classification, source-page citations, and conflict flags — for Investment Banking Analysts and PE Associates populating deal screening models.
Skill Claude CodeCodex
Given a borrower's financial spread, credit agreement term sheet, and industry sector, drafts each named narrative section of a commercial credit memo — company overview, management, industry analysis, financial performance, credit strengths and risks, mitigants, and recommendation — with citations to input data and…
Skill Claude CodeCodex
Given a current earnings call transcript and optionally a prior quarter's transcript, extracts and categorizes every forward-guidance statement by confidence level, flags management tone shifts quarter-over-quarter using an explicit hedging-language taxonomy, and produces a structured analyst brief distinguishing…
Skill Claude CodeCodex
Audits an AI-generated LBO model for formula logic errors, benchmarks assumptions against sector norms, generates a stress-test scenario matrix, and produces a firm-convention compliance checklist — for PE and IB associates doing final review before a model is presentation-ready.
Skill Claude CodeCodex
Given a credit agreement and a borrower's quarterly financial submission, extracts contract-specific covenant definitions, calculates each required ratio from submitted financials, and produces a compliance certificate with pass/fail status and breach-proximity alerts — for Credit Analysts and Portfolio Managers.
Skill Claude CodeCodex
Given a completed comparable companies table and target company description, drafts the full interpretive narrative for an investment banking pitchbook comps page — peer set rationale, outlier analysis, target positioning, and slide-ready callout bullets.