DottaandPaperclip
433d94ee0b
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Feature recipes and Storybook help contributors find behavior, examples, and verification steps. > - Independent feature changes currently update shared catalogs, indexes, and review hashes. > - Those shared files cause merge conflicts and repeated bookkeeping during parallel work. > - This pull request stores metadata in each feature recipe and story, then generates the combined views. > - Contributors can merge independent work without editing a common index. Git and PR evidence retain change and review history. ## Linked Issues or Issue Description **What existing behavior does this improve?** Maintaining the product, agent-access, and Storybook maps while multiple agents change different features. **Current behavior** A story or recipe change can update `catalog.json`, `INDEX.md`, review hashes, verification tables, and tracked generated navigation. Two unrelated changes then conflict in the same files. **Proposed behavior** Each story declares capability links and classification in `parameters.paperclip`. Each feature recipe declares its navigation and optional verification guidance. Startup, build, and UI typechecking discover these sources and generate ignored indexes and Storybook pages. **Reason and benefit** A routine change edits its own source files. Aggregate metadata is rebuilt from the merged checkout. There is no sidecar registry, shared review journal, or hash acknowledgment to maintain. **Breaking changes** The tracked Storybook catalog and index paths are retired. Consumers use `ui/storybook/generated/` or `maps/generated/` after generation. Contributor instructions and the existing feature-gap browser suite use the new paths. Story and overview IDs remain stable. ## What Changed - Move the existing annotations into 262 story files. Preserve classifications, capability associations, exports, and explicit IDs. - Discover all 35 feature recipes. Store section/order beside each recipe and move durable verification routes out of the shared table. - Remove tracked catalogs, inventories, overview pages, and inline preview ordering. Keep authored preview behavior in `preview-base.tsx`. - Generate the Storybook guide, feature/access Markdown indexes, gap report, page-reference inventory, and preview wrapper automatically. - Update the map contracts and maintenance skills to use Git history and revision-specific PR/CI evidence. - Add schema, identity, regeneration, and actual Git merge tests. Run these through the existing UI Vitest suite. Add focused browser navigation checks. This is one coordinated migration. Most of the 367 changed files contain only moved metadata or deleted generated content. Review the generator scripts, map contracts, Storybook configuration, and tests first. No product API or authorization behavior changes. ## Verification Passed: - `node scripts/check-maps.mjs`: 35 paired areas and 175 capabilities. - `node scripts/storybook-catalog.mjs --check`: 262 suites and 2,485 authored stories after incorporating current master. - `node --test scripts/check-maps.test.mjs scripts/storybook-catalog.test.mjs scripts/storybook-feature-index.test.mjs`: 28 tests, including independent branches merged in both orders with identical generated output and no tracked changes. - `pnpm --filter @paperclipai/ui exec vitest run storybook/map-tooling.test.ts`: the CI entrypoint runs those map tests. - `pnpm check:token-gates`. - `pnpm -r typecheck` and `pnpm build` before the final master rebase. - `pnpm build-storybook`: all 2,517 original story/docs URLs retained. Current master contributes exactly four additional recovery stories. - Focused Playwright navigation checks: search and open a feature without examples; expand verification guidance and check source links; retain capability order and follow a related Foundations suite. The final clean Storybook build passed after deleting every ignored map output. Post-rebase UI typechecking and both browser tests passed. The working tree stays clean after generation. The concurrent-cleanup regression fails with `ENOENT` without the fix and passes with it. All GitHub CI gates passed on `00fc7ec18e`, which also received Greptile 5/5 with no unresolved findings. The final merge commit `1c3c5143a6` incorporates master and resolves only the two modify/delete conflicts by retaining removal of the aggregate index and catalog. Master’s new workspace feature and access guidance is preserved. Map validation, catalog generation, overview generation, all 28 map tests, and diff checks pass after conflict resolution. The maintainer explicitly requested merging without another Greptile review. CI has not been rerun to completion on this merge commit. The local repository-wide `pnpm test:run` was stopped before completion after the earlier CI matrix passed; it is not reported as a local pass. No screenshot baselines or production acceptance are claimed. To repeat the browser check against an owned Storybook server: ```sh PAPERCLIP_GAP_STORYBOOK_URL=http://127.0.0.1:6136 pnpm exec playwright test --config tests/storybook-visual/feature-gap-stories.config.ts map-navigation.spec.ts ``` ## Risks - Existing branches need a one-time rebase to move their catalog changes into local annotations. The PR incorporates the recovery changes from current master. - Changing the same story or capability can still produce a meaningful conflict. This removes the shared aggregate bottleneck. - The generator uses Storybook's existing CSF parser. Literal-only annotations and parser regression tests protect the current contract. - Generated Markdown is available after running the generator; the published Storybook provides a browsable guide. - Shared source changes still require review. Removing hashes removes mechanical acknowledgments; it does not establish runtime parity. ## Model Used OpenAI Codex, based on GPT-6, with repository inspection, code execution, and browser-test tools. The exact deployment model ID and context window were not exposed in this session. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green on the previously reviewed head `00fc7ec18e`; final conflict-resolution checks are documented above - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups on `00fc7ec18e`; another review was waived by the maintainer - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>
6.5 KiB
navigation
| navigation | ||||||
|---|---|---|---|---|---|---|
|
CLI/API and local worktrees
Operators and agents can work through explicit CLI/API context, manage domain resources, inspect runs, and create isolated local worktree instances for development.
Implementation: command registry, client context, worktree commands.
Agent access: companion.
Sub-features
context: choose instance/API base, profile, identity, and company explicitly.domain-commands: read/write tasks, agents, projects, goals, approvals, routines, and other registered resources.outputs: work with documents/assets/skills and inspect machine-readable results.run-controls: invoke/inspect supported runs and observe actual outcomes.worktree-instances: initialize, seed, diagnose, and clean up only an owned development instance.
How to get to it (user POV)
client-context
Use paperclipai context show/list/use/set, auth/connect, and the registered resource command help.
subresources-and-runs
Use issue subresource, asset/skill, and run/heartbeat commands exposed by the CLI help.
worktree-instance
Use the documented worktree/test-drive commands in an owned checkout; inspect paperclipai worktree --help first.
Inspect root help too: worktree:make/list/merge-history/cleanup are top-level
commands rather than subcommands of worktree.
Driving it with the Paperclip CLI
Preconditions: follow the baseline. Use --help from the checked-out CLI to discover exact arguments. Select a disposable company and isolated context; do not inherit an unknown API target.
Pass the server origin to --api-base (without appending /api); the commands
already include the API path. issue create requires explicit --company-id
even when a profile stores a company. A persona label is not authentication proof.
client-context
Automated: context, HTTP client, and operations parity cover client contracts, not every live endpoint.
Manual: Inspect the selected context and identity, list the disposable company’s resources, create a harmless task through the supported command, and verify it in the browser. Exercise an invalid token and wrong-company resource. Inspect JSON output and nonzero errors rather than only printed success text.
subresources-and-runs
Automated: issue subresources and asset/skill parity cover command/API contracts.
Manual: Write a small task document or attach a fixture using the supported command, open it in the UI, and compare contents. Invoke eligible test work, inspect the resulting run ID/status/output, and verify denial when the actor lacks permission.
worktree-instance
Automated: worktree commands and test drive cover isolation/setup contracts.
Manual: Initialize an owned disposable worktree instance, inspect its printed paths and ports, verify health and login, and confirm any seed data belongs to the intended source. Stop and clean up only that instance; inspect preserved work before deleting any checkout.
Gotchas
- API success proves its contract, not the equivalent browser creation journey.
- The CLI command registry is broader than the UI page inventory; new command families need a recipe review.
- See instance operations for install/update/backup and feature-specific recipes for domain behavior.
Storybook verification
context
Scope: cli Storybook: No authored story required for CLI profile/context selection. Production: Verify selected API origin, identity and company, JSON output and invalid-token/wrong-company failures. A persona label is not authentication. Sources: Source: context.ts. Tests: context.test.ts, http.test.ts
domain-commands
Scope: cli Storybook: No authored story required for command parsing and domain API execution; corresponding product UI keeps its own story needs. Production: Discover registered help, use a disposable company, assert persisted mutations and denied operations with the actual actor. Command parity is not every endpoint passing live. Sources: Source: index.ts, issue.ts. Tests: operations-parity.test.ts
outputs
Scope: cli Storybook: No authored story required for CLI document/asset/skill IO or machine-readable output. Production: Assert payload/schema, exit code, written contents, authorization and file integrity. Open the resulting document/asset in the product for cross-surface acceptance. Sources: Source: issue.ts, asset.ts, skill.ts. Tests: issue-subresources.test.ts, admin-asset-skill-parity.test.ts
run-controls
Scope: cli Storybook: No authored story required for invoke/list/get/cancel/events/log command execution. Production: Assert the real run ID, event/log output and resulting state, eligible actor and cancellation denial. Printed success does not prove provider completion. Sources: Source: run.ts, heartbeat-run.ts. Tests: operations-parity.test.ts
worktree-instances
Scope: cli Storybook: No authored story required for worktree initialize/seed/repair/cleanup commands. Production: Use only an owned isolated checkout/instance; verify paths, ports, health, login and seed provenance, then cleanup ownership. Capture evidence before teardown. Sources: Source: worktree.ts, test-drive.ts. Tests: worktree.test.ts, test-drive.test.ts