Open source framework for spec-driven agentic software engineering. PRDs, prompts, agents, skills, workflows, and coding rules for Codex, Claude, and modern AI development teams.
Agent Claude Code
This directory contains the 12 Claude Code agents that implement the SpecRoute skeleton. They are project-internal - they build SpecRoute itself, not consumer-facing templates that ship as artifacts.
Agent Claude Code
Use when designing or extending the consumer-facing agent roster under agents/. Owns agents/agent-template.md (frontmatter contract), agents/archetypes/ (role concepts - product, architect, backend, frontend, security, qa, devops), agents/examples/ (flat .md files showing real frontmatter shape), and agents/roster.md…
Agent Claude Code
Use when drafting or reviewing slash commands under commands/. Owns vendor-specific command templates - command-template.claude.md (markdown body + frontmatter), command-template.gemini.toml (TOML command file with prompt + description). Documents Codex's "skill with user-invocable - true" approach. Ensures per-vendor…
Agent Claude Code
Use when checking whether SpecRoute's claims about the outside world are still true - vendor config surfaces, hook event taxonomies, frontmatter contracts, version anchors, transition dates, and counts that must match disk. Owns the currency cycle, not the prose. Runs before releases, after any vendor ships a breaking…
Agent Claude Code
Use when drafting or revising framework documentation under agentic-docs/. Owns philosophy.md, spec-driven-development.md, agentic-coding-model.md, automation-decision-framework.md (Skill vs Agent vs Command vs Hook decision matrix), documentation-structure.md, two-tier-docs-pattern.md, multi-vendor-context-files.md…
Agent Claude Code
Use when drafting or reviewing hooks under hooks/. Owns the per-vendor starter templates and scripts, including Claude's 3-event starter and Kiro v1 JSON with 10 triggers. Documents each vendor's native taxonomy and blocking contract. Triggers - "draft the hooks.json template", "add a Kiro file-pattern hook", "what…
Agent Claude Code
Use when drafting, reviewing, or extending PRD content under prds/. Owns the full 23-section enterprise PRD template, the lightweight PRD template, the platform-SRS template, the PRD lifecycle directories (active/deprecated/archive), and example PRDs. Triggers - "draft a PRD template", "fill out prd-template.md"…
Agent Claude Code
Use when drafting or reviewing prompts under prompts/. Owns the master/phase/task prompt template trio in prompts/shared/ (global-master-prompt-template.md, phase-master-prompt-template.md, task-prompt-template.md), per-vendor prompts in prompts/codex/ and prompts/claude/, and shared prompts (prd-to-spec…
Agent Claude CodeCodex
Use when building or updating runtime layouts under runtimes/, the MCP single-source-of-truth, or cross-vendor sync utilities under tools/. Owns the six copy-pasteable layouts at runtimes/.claude/, runtimes/.codex/, runtimes/.gemini/, runtimes/.kiro/, runtimes/.cursor/, and runtimes/.devin/, plus runtimes/mcp/…
Agent Claude Code
Use proactively before any commit and whenever new content is added. Scans tracked files for private-project leaks - upstream private project names, internal absolute paths, proprietary domain logic (financial / portfolio / trading / prediction / tax / advisory specifics), customer data, secrets, internal endpoints…
Agent Claude Code
Use when designing or extending skills under skills/. Owns the folder-per-skill convention with SKILL.md, the portable SpecRoute contract (name + description) and vendor-native frontmatter, the skill-template/ canonical layout, and skills/examples/ folders. Ensures skills are interactive parameterized workflows with a…
Agent Claude Code
Use when drafting or reviewing spec content under specs/. Owns the spec triplet (requirements-template.md + design-template.md + tasks-template.md), the lightweight feature-spec-template.md, the technical-spec-template.md, and architecture-decision-record.md. Ensures requirement IDs are stable and tasks back-reference…
Agent Claude Code
Use when reviewing templates for the "production-grade and immediately usable" criterion. Rejects abstract checklists, theory-only docs, and templates that don't produce valid artifacts when filled in. Verifies frontmatter contracts (skills, agents, commands), checks that examples exist alongside templates, ensures…
Agent Claude Code
. Owns . Triggers - 'literal user utterance 1', 'literal user utterance 2', 'literal user utterance 3'." Trigger phrases drive auto-selection - without them the agent will not be invoked.>.
Agent Claude Code
Cross-vendor inventory of agents for a SpecRoute-driven project. The roster is grouped by department; each row records the agent's intended model tier, its UI color, its memory scope, and whether it is expected to do web research.
Agent Claude Code
General-purpose backend implementer for the user-search feature. Owns API handlers, request validation, RBAC scoping, query builders, cache logic, observability instrumentation, and rate limiting. Triggers - "implement task.
Agent Claude Code
Database engineer for the user-search feature. Owns the index migration, query plan validation, and schema decisions. Triggers - "run task.
Agent Claude Code
Deployment / SRE for the user-search feature. Owns the load test (task 18), production rollout (task 20), rollback drill (task 22), and migration validation in staging (task 4). Triggers - "run the load test against staging", "begin the 10% production rollout", "execute the rollback drill", "validate the index…
Agent Claude Code
Frontend implementer for the user-search feature. Owns the SearchInput, FilterChips, ResultTable, CursorPagination components and the /users/search page composition. Triggers - "implement task.
Agent Claude Code
Integration / E2E test author for the user-search feature. Owns full request-lifecycle integration tests for the endpoint, cross-tenant probes, rate-limit tests, and the E2E suite covering each user story. Triggers - "write integration tests for task.
Agent Claude Code
Security reviewer for the user-search feature. Reviews tasks 6 (RBAC scoping), 8 (cursor encoding), 11 (audit log content), 12 (rate limiting). Owns the privacy review for query-string logging (task 3). Triggers - "review the RBAC boundary", "audit the cursor signing", "review the audit log shape", "validate the…
Agent Claude Code
Unit test author for the user-search feature. Writes tests against requirement IDs back-referenced by tasks - validator rules, RBAC matrix, query-builder SQL shapes, cursor round-trips, cache hit/miss paths, observability metrics, frontend component behaviors. Triggers - "write unit tests for task.
Agent Claude Code
Use when reviewing a PRD, requirements/design/tasks triplet, implementation plan, or agent prompt for completeness before code work starts. Owns requirement coverage, acceptance criteria quality, task back-references, and unresolved open questions. Triggers - "review this spec", "audit the requirements", "check this…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: