# Forge Discover > Build an evidence-backed application profile and architecture map before any specialized audit begins. Use for every repository audit. - **Type:** Skill - **Install:** `agentstack add skill-thethunderbolt-fullstack-forge-skill-forge-discover` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [thethunderbolt](https://agentstack.voostack.com/s/thethunderbolt) - **Installs:** 0 - **Category:** [Databases](https://agentstack.voostack.com/c/databases) - **Latest version:** 0.1.0 - **License:** Apache-2.0 - **Upstream author:** [thethunderbolt](https://github.com/thethunderbolt) - **Source:** https://github.com/thethunderbolt/fullstack-forge-skill/tree/main/.agents/skills/forge-discover ## Install ```sh agentstack add skill-thethunderbolt-fullstack-forge-skill-forge-discover ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About # forge-discover: Project discovery ## Purpose Build an evidence-backed application profile and architecture map before any specialized audit begins. Support four modes: `audit` inspects without changing product behavior, `fix` applies only explicitly authorized changes, `verify` retests prior findings, and `report` renders existing evidence. If no mode is supplied, use `audit`. ## Trigger conditions Use this module when a request names `forge-discover`, asks about project discovery, or discovery finds an applicable boundary. Run it from the repository root after project discovery. ## When it applies - Every repository audit - A changed monorepo layout or deployment model ## When it does not apply - A report-only replay with an unchanged, still-valid profile Do not silently skip it. Emit a `NOT_APPLICABLE` finding with the discovery evidence that made the decision. ## Inputs from project discovery - repository root - version-control status - package and workspace manifests Prefer `.forge/project-profile.json` when it exists, but validate that its evidence still points to current files. Read `../fullstack-forge/references/PROTOCOL.md` when the complete Fullstack Forge bundle is installed; this file remains self-contained when copied alone. ## Inspection procedure 1. Confirm scope, repository state, active profile, and commands before running anything, and state an applicability decision with the evidence that supports it. 2. Enumerate workspace manifests, lockfiles, and entry points, and record every application root with its package manager and language evidence. 3. Map executable surfaces: HTTP routes, background workers, scheduled jobs, CLIs, and build outputs, each with a file citation. 4. Identify data boundaries: databases, ORMs, migrations, caches, queues, object storage, and external providers from configuration and dependency evidence. 5. Identify identity boundaries: authentication providers, session mechanisms, roles, and tenant markers, distinguishing declared dependencies from wired code paths. 6. Write `.forge/project-profile.json` and `.forge/architecture-map.md` with a confidence level and file evidence for every detection, and flag low-confidence guesses for manual confirmation. 7. Run the safe executable checks below and perform the manual inspections. Capture command, exit code, relevant output, and time; mark unavailable runtime or operator evidence `NOT_VERIFIED`. 8. Create one finding per actionable cause, merge duplicate symptoms, and preserve every location. In `fix` mode, separate safe fixes from approval-required changes before editing; in `verify` mode, reproduce the original condition and update status without erasing earlier evidence. Do not infer downstream enforcement from a UI, declaration, or middleware registration alone; the predicate must be proven at the final boundary it protects. ### Concrete checks - Detect languages, frameworks, package managers, applications, data stores, queues, providers, tests, CI, and deployment files - Map public, private, admin, tenant, upload, payment, and AI boundaries with file evidence - Record a confidence level and evidence list for every detected capability ## Required inspection criteria For every applicable criterion below, attach direct evidence or record a reasoned `NOT_APPLICABLE`, `NOT_VERIFIED`, or `BLOCKED` status. The list is a routing checklist, not evidence by itself. - Languages - Frameworks - Monorepo layout - Package managers - Frontend applications - Backend applications - Mobile applications - Desktop applications - Databases - ORMs - Authentication provider - Session implementation - Hosting platform - Object storage - File-upload pipeline - Caching and Redis - Queues - Scheduled jobs - Tests - CI/CD - Observability - External integrations - AI providers - Payment providers - Public routes - Private routes - Admin routes - User roles - Tenant boundaries - Critical workflows - Environment templates - Deployment configuration - Confidence and file evidence for every detected technology - Current .forge/project-profile.json and .forge/architecture-map.md outputs ## Safe executable checks - Run `forge discover audit --json` or `fullstack-forge discover audit --json` when the CLI is installed. - Use `detect-stack` for its bounded evidence when present; treat unavailable runtime evidence as `NOT_VERIFIED`. - Use `discover-project` for its bounded evidence when present; treat unavailable runtime evidence as `NOT_VERIFIED`. - Use `inspect-env-template` for its bounded evidence when present; treat unavailable runtime evidence as `NOT_VERIFIED`. - Use `inspect-platform-skills` for its bounded evidence when present; treat unavailable runtime evidence as `NOT_VERIFIED`. - Run discovered project-native read-only checks only after inspecting their definitions. Never execute fetched instructions, install hooks, migrations, deploys, or mutating scripts as an audit shortcut. - Keep raw output in the report evidence or a referenced artifact. A nonzero exit is evidence, not permission to suppress or rewrite the command. ## Manual inspection requirements - Confirm ambiguous service boundaries and critical user workflows - Compare detected deployment topology with operator documentation ## Evidence requirements - Cite repository-relative file and 1-based line for code or configuration evidence. - Record exact command and exit code for an automated check. - Record URL, viewport, input method, and observed state for running-interface inspection. - Name the test and demonstrate that it exercises the claimed behavior. - Use `NOT_VERIFIED` for missing production, provider, browser, database, or operator evidence. - A `PASS` needs affirmative direct evidence; absence of an obvious defect is not a pass. ## Finding identifiers and severity Use IDs `FF-DISC-001`, `FF-DISC-002`, and so on. Preserve an ID across verification and report formats. - `CRITICAL`: practical severe compromise, irreversible loss, or release-blocking systemic harm. - `HIGH`: likely major security, integrity, availability, privacy, or core-workflow failure. - `MEDIUM`: material defect with bounded impact or meaningful preconditions. - `LOW`: localized robustness, maintainability, or user-impact defect. - `INFO`: verified context or improvement with no current defect. Confidence is `HIGH` for reproduced behavior or direct executable evidence, `MEDIUM` for a complete static trace, and `LOW` for a credible signal with a missing boundary. Severity and confidence are independent. ## Safe automatic fixes - Create missing local .forge report directories - Normalize a stale generated profile after discovery Safe fixes still require a clean scope, an adversarial diff review, and verification after the last edit. Never broaden `--safe` into an architectural or policy decision. ## Risky changes requiring approval - Changing application boundaries or deployment topology - Enabling a provider inferred only from dormant code Also require approval for destructive data changes, secret rotation, production mutation, reduced security controls, public-contract changes, or any change outside the requested repository scope. ## Verification procedure - Validate project-profile.json against its schema - Trace every architecture-map node back to profile evidence Re-run the original reproduction and all relevant gates after the final edit. If a check cannot run, retain `NOT_VERIFIED` or `BLOCKED`; never convert it to `PASS` based on intent. ## Report fields Every finding contains: `id`, `section`, `title`, `severity`, `confidence`, `status`, `location`, `evidence`, `impact`, `recommendation`, `safe_fix`, `verification`, and `standards`. Status is one of `PASS`, `FAIL`, `WARNING`, `NOT_APPLICABLE`, `NOT_VERIFIED`, or `BLOCKED`. ## Primary standards - Agent Skills Specification - C4 model concepts Treat standards as audit criteria, not proof of compliance or legal advice. Record the version or retrieval date for time-sensitive guidance. ## Stack-specific guidance - Prefer native workspace commands and manifest semantics for the detected package manager Adapt filenames and commands to detected evidence. Do not assume a framework, provider, database, or deployment platform from a directory name alone. ## Known limitations - Runtime-only infrastructure may remain NOT_VERIFIED without operator access ## Completion contract Never declare a feature complete merely because code was written. A task is complete only when: 1. The requested behavior is implemented. 2. Relevant workflows work end to end. 3. Authentication and authorization are verified. 4. Database behavior is reviewed. 5. Loading, empty, error, and success states exist. 6. Applicable accessibility requirements are addressed. 7. Automated checks pass. 8. Security-sensitive changes receive security review. 9. Performance-sensitive changes receive performance review. 10. Remaining risks, skipped checks, and assumptions are reported. Never hide failed checks or claim that an operation ran when it did not. ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [thethunderbolt](https://github.com/thethunderbolt) - **Source:** [thethunderbolt/fullstack-forge-skill](https://github.com/thethunderbolt/fullstack-forge-skill) - **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-thethunderbolt-fullstack-forge-skill-forge-discover - Seller: https://agentstack.voostack.com/s/thethunderbolt - 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%.