Anthony Ettinger
9ae7ad8962
* OpenFleet reference implementation: @logicsrc/openfleet 0.1.0 and logicsrc fleet Ship what docs/openfleet.md describes. The new workspace package holds the record (write once, never overwrite, 0600), the ledger (append-only JSON Lines, merged across ledger*.jsonl by at), the ceiling rules (whole fleet ceiling, narrowed swarm keys, a merge that never widens, refusals by key), claiming and deriving exactly as the spec's "Claiming and deriving" and rule 13, and fold(), which turns any $OPENFLEET_HOME plus the engine rosters into the tree the landing page shows. logicsrc fleet open|cap|tree|stop|log are the sysop's verbs, every one with --json. open and cap exit 4 when OPENFLEET_MEMBER is set; stop exits 4 outside the caller's subtree, ends nested swarms first, goes through each member's own engine (claude stop, moshcode herd kill, tmux kill-pane, a signal for claude-p) and writes one swarm.end per swarm. tree reads claude agents --json --all and ~/.moshcode/herd/sessions.json when it can, draws recordless sessions as roster roots of the implicit fleet, and writes member.end lost for a recorded member its engine no longer lists. Claude Code takes part through hooks: logicsrc fleet hooks install merges SessionStart, UserPromptSubmit, PreToolUse, Stop and SessionEnd into ~/.claude/settings.json without clobbering it, and logicsrc fleet hook <Event> runs each one. SessionStart claims, derives or writes a root record and hands the member its variables through CLAUDE_ENV_FILE; UserPromptSubmit checks the ceiling with the permission mode the engine reports and writes member.start, or refuses the first prompt with exit 2 and ceiling.refuse; PreToolUse denies an edit outside piece.owns; Stop and SessionEnd write member.end. A hand-started root takes the engine's reported approvals before member.start, since the command line only guesses them. Hooks never fail the engine: everything is caught and logged to hooks.log. The spec and the landing page now say what ships, keep Status 0.1, and record the two verified Claude Code limits: a background job dispatched from claude agents gets no launcher environment, and OPENFLEET_* exported at SessionStart reach the member's tools but not later hooks, so hooks key on session_id through $OPENFLEET_HOME/sessions/<session_id>.json. PRD 0008 covers the work. CLI 0.2.1 -> 0.3.0; build and build:cli chains build the package before the CLI; README and docs/cli.md list the group. Tests: 95 in the package (record, ledger merge, every narrower case, the worked example's claim and derive, the folded tree, hook install idempotence, each hook handler including the exit-2 refusal and the PreToolUse deny, every verb with fake deps) and 4 in the CLI. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RZV4zJ2pDZLNN3kE5jFCmV * OpenFleet fix round: rebuild the ceiling from the ledger, once-markers, rule 6 in tree, lost only for what a roster can hold The review of the reference implementation against moshcode found the two readers disagreeing on the same files. This round applies the shared rulings so both sides read a ledger the same way. Ceiling (R-A, R-B, R-C, R1, R6, R10, R15, R17): memberCeiling rebuilds the effective ceiling from the ledger on every read. The latest fleet-target fleet.cap (else fleet.open, else the implicit fleet's) replaces the copy in a record, so a sysop's widening cap reaches running members; then each swarm.spawn narrowing down the path, then swarm caps last. In the implicit fleet a parentless record's own approvals enters at the root; a ceiling a writer left without the key is never read as native, and startMember fills it with the engine's word while the record is unclaimed. A fleet.open or cap with no hosts means the host it was written on (R23). Once-markers (R-G, R28): member.start, member.end and swarm.end each take an exclusive create under fleets/<fleet>/marks/<event>.<id> before the append; a lost end takes <id>.lost so a real end can still supersede it. The hooks let a real end follow a lost line (R9). tree (R-F, R20): run by the sysop it enforces rule 6, stopping a member past its effective until with state timeout and the members of a swarm or fleet at its budget with state budget, then writes swarm.end for each swarm touched once it is complete. An agent's tree stops nothing. lost is written only for a member its engine's roster can hold: a claude-code background job (8-hex member or session) or a moshcode pane, never an interactive session claude agents does not list (R-E, R3, R14). A nested swarm is drawn under the member that spawned it and its row shows the effective ceiling (R25). stop and cap (R-D, R-H, R22, R27): swarm.end is written only once every member and every nested swarm has an end line that counts; an engine that will not end a member leaves it without an end line and the verb exits non-zero. claude stop takes the job id: the member of a background job, else the first eight characters of a session UUID; an interactive session with no job id cannot be stopped and the tool says so. cap on a swarm refuses a key that would widen. A derived claude-code job is named by its job id and carries no pid. Also: R-I (endMember ends only the engine-minted swarm of one), R35 (a derived record's guessed approvals corrected at UserPromptSubmit), R32 (the UserPromptSubmit hook passes only exit 2 through), R31 (package README), R36 (rule 13 says the launcher test is unimplemented in 0.1), docs and PRD 0008 updated for lost, rule 6 and the markers. 113 openfleet tests, 93 CLI tests, contract green. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RZV4zJ2pDZLNN3kE5jFCmV * openfleet hooks: no member.end for a member that never started A first prompt refused by the ceiling still lets the session wind down through Stop and SessionEnd; those handlers now write nothing when the ledger holds no member.start for the member, so a refused member is never drawn as done. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RZV4zJ2pDZLNN3kE5jFCmV * logicsrc-mcp test: the next free PRD id is 0009 now that PRD 0008 exists Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01RZV4zJ2pDZLNN3kE5jFCmV --------- Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com> |
||
|---|---|---|
| .github | ||
| apps | ||
| docs | ||
| examples | ||
| packages | ||
| plugins | ||
| prd | ||
| scripts | ||
| supabase | ||
| .env.example | ||
| .gitignore | ||
| LICENSE | ||
| package-lock.json | ||
| package.json | ||
| README.md | ||
| tsconfig.base.json | ||
LogicSRC
LogicSRC is an open standards initiative for human and AI agent coordination, maintained by Profullstack, Inc.
CommandBoard.run is the first hosted product built on LogicSRC: a modern BBS where humans and AI agents coordinate work through boards, tasks, DID identity, OAuth, CLI, TUI, plugins, reputation, audit logs, and payments.
The standards surface is named logicsrc. External tools can consume LogicSRC contracts, but the LogicSRC CLI remains the OpenStandards command surface.
Monorepo
apps/
commandboard-api REST API reference service
commandboard-web PWA shell
packages/
cli logicsrc OpenSpec CLI
opencontext OpenContext reference implementation (resolver, scopes, bundles, adapters)
openontology OpenOntology reference engine (entities, claims, queries, change sets)
openprd OpenPRD reference implementation (numbered PRDs, lifecycle, task bridge)
openfleet OpenFleet reference implementation (record, ledger, sysop verbs, Claude Code hooks)
logicsrc-mcp @profullstack/logicsrc-mcp standards MCP server
sdk SDK contract types and helpers
tui terminal UI
schemas LogicSRC JSON schemas
validators schema validation utilities
agentad AgentAd Marketplace exchange (auction, metering, settlement)
plugin-core plugin manifest and loader runtime
plugins/
coinpay default DID, wallet, payment, and escrow plugin
ugig default jobs and gigs marketplace plugin
c0mpute work-in-progress compute jobs and worker pools plugin
docs/
specs, CLI conventions, permissions, and roadmap notes
examples/
openontology/ethereum-ecosystem fictional ecosystem map demonstrating OpenOntology
opencontext/* five OpenContext repositories, from minimal to multi-agent
prd/
numbered OpenPRD proposals
scripts/
install.sh curl | sh installer
Quick Start
npm install
npm run check
npm --workspace @logicsrc/cli run dev -- --openspec agentswarm --yolo --repo profullstack/logicsrc
npm --workspace @logicsrc/cli run dev -- openspec import
npm --workspace @logicsrc/cli run dev -- openspec export --out logicsrc-openspec-summary.md
npm --workspace @logicsrc/cli run dev -- --openspec-only task validate packages/schemas/fixtures/task.yaml
npm --workspace @logicsrc/cli run dev -- agentswarm --yolo --repo profullstack/logicsrc
npm --workspace @logicsrc/cli run dev -- plugins
npm --workspace @logicsrc/cli run dev -- tui
npm --workspace @profullstack/logicsrc-mcp run build
node packages/logicsrc-mcp/dist/index.js
OpenWall proposal
OpenWall proposes broadcasts to explicitly selected connections, followers,
following, or service users, plus direct messages through verified contact routes. The draft
defines recipient consent, private delivery outcomes, and a public/private AT Protocol mapping.
Message and receipt JSON Schemas and fixtures are included; delivery services and the proposed
logicsrc wall commands are not implemented.
OpenPRD
OpenPRD is a lightweight standard for product requirements documents: a repo
keeps a numbered, committed collection under prd/, one Markdown file each, with front-matter, a
fixed set of eight sections, and a lifecycle. @logicsrc/openprd implements it.
npm --workspace @logicsrc/cli run dev -- prd new "Expand the parked-domain service"
npm --workspace @logicsrc/cli run dev -- prd validate ./prd --strict
npm --workspace @logicsrc/cli run dev -- prd status 0001 Review
npm --workspace @logicsrc/cli run dev -- prd tasks 0001 --priority P0
Conformance failures (filename, front-matter, id match, the eight sections in order) are errors;
lint findings are warnings that --strict promotes. The lifecycle is enforced — Draft cannot
jump to Final, and Superseded must name its replacement. prd tasks is the optional bridge:
each R# becomes one schema-valid logicsrc.task.
OpenOntology
LogicSRC OpenOntology is an open contract for durable, source-backed domain knowledge shared by humans and AI agents: typed entities, claims that carry provenance and time, a portable query AST, and governed change sets. It is storage-agnostic, model-provider-neutral, and works offline with no account.
npm --workspace @logicsrc/cli run dev -- ontology init my-ecosystem
npm --workspace @logicsrc/cli run dev -- ontology validate my-ecosystem --strict
npm --workspace @logicsrc/cli run dev -- ontology query run contributors --dir my-ecosystem
✓ 3 entity types
✓ 4 relationship types
✓ 8 entities
✓ 14 claims
✓ 2 sources
OpenOntology package is valid.
Claims are append-only and agents propose rather than apply: a corrected fact becomes a dispute, retraction, or supersession, and every answer traces back to the claims, evidence, and sources behind it.
Surfaces: a SQLite/Turso storage adapter, a REST + SSE reference service described by OpenAPI at
/api/ontologies/openapi, MCP resources and tools, JSON-LD/RDF/SHACL export, seven source adapters
that propose rather than apply, keyboard-first TUI panels, and a read-only web explorer at
/openontology/explore. See also
governance and
interoperability.
OpenContext
LogicSRC OpenContext is an open specification for durable, portable, permissioned, provenance-aware context shared between humans and AI agents. It defines how organizational knowledge is described, authorized, versioned, resolved, audited, and handed between replaceable workers without losing institutional state.
An agent should be replaceable without losing organizational knowledge.
npx @logicsrc/opencontext init my-context
npx @logicsrc/opencontext validate --strict
npx @logicsrc/opencontext resolve --role support --task "customer asked for a refund" --explain
Included:
✓ mission canonical
✓ policies.refunds canonical
✓ procedures.refund approved
Excluded:
- decisions.2026-08-09-adopt-opencontext not-in-scope (no include pattern matches)
Digest: sha256:81b41a915ee68f744e91ef0d7760440de51b603088de1a6f21ea6f337bb374a8
Authorization runs before relevance, so an agent never ranks context it may not read; authority is declared rather than inferred from retrieval rank; unresolved canonical conflicts are reported rather than quietly settled; and resolution is deterministic, so a decision can cite the exact bundle digest it was made from. Untrusted content — tickets, chats, scraped pages — keeps its trust level through resolution and is fenced and labelled in rendered bundles.
It is not a memory database. Memory is one possible context source; OpenContext is the control plane above systems that remain the sources of truth. It runs from a folder and a Git repository with no account, no server, and no telemetry.
Also available as logicsrc context <command>, sharing one implementation with the standalone
binary. See the specification, CLI,
SDK, security model, and
conformance guide.
One command for every LogicSRC tool
logicsrc is the umbrella: every LogicSRC standard that has a CLI is a word after it, and the word runs the same code as the standalone command, so the two cannot drift.
logicsrc vault … # OpenCreds (also `opencreds`)
logicsrc prd … # OpenPRD
logicsrc ontology … # OpenOntology
logicsrc context … # OpenContext (also `opencontext`)
logicsrc fleet … # OpenFleet: open, cap, tree, stop, log, hooks install
logicsrc openmcp … # OpenMCP: relays, find, call, add, probe, serve (also `openmcp`)
logicsrc openspec … # import, export, change; any other word is OpenSpec.dev's own CLI (init, list, validate, archive, show)
logicsrc mcp # the LogicSRC MCP server over stdio (also `logicsrc-mcp`)
openmcp is the one that lives in its own repository (logicsrc/openmcp); it is a dependency here and its main is called with your arguments untouched. It needs Node 24 (node:sqlite); on an older Node that one word says so and the rest of the CLI keeps working. The standalone install, which brings its own Node 24, is curl -fsSL https://openmcp.logicsrc.com/install.sh | sh.
MCP
LogicSRC exposes a standards-focused MCP server as @profullstack/logicsrc-mcp, and logicsrc mcp runs it.
It provides read-only resources for docs and schemas, validation/example tools, and prompt templates for creating LogicSRC-compatible documents.
v1.0.0 Priorities
- LogicSRC task, agent, run, event, permission, and plugin schemas.
- AgentAd: disclosed, agent-readable ad schemas for CLI/agent advertising (see
docs/agentad.md); cl1s.tech is the reference network. The two-sided exchange on top is specified indocs/agentad-marketplace.md. - LogicSRC CLI, SDK, TUI, PWA, MCP, and curl-compatible API conventions.
- CommandBoard.run reference implementation.
- Monorepo-maintained plugin system.
- Credential Sharing OpenSpec for end-to-end-encrypted team vaults, .env, Doppler, Railway variables, GitHub Secrets, sh1pt, and
~/.sshkeys. - CoinPay as the default payment, DID, wallet, and escrow plugin.
- uGig as the default jobs and gigs marketplace plugin.
- c0mpute as a work-in-progress compute jobs and worker pools plugin.
- Installer, update/upgrade, remove/uninstall workflows.
OpenABTest draft
OpenABTest defines reusable experiment manifests and private eligibility, assignment, exposure, conversion and accounting events. The Chovy example compares 5%, 10% and 20% discounts on every referred purchase, with sticky customer assignment and affiliate payout withheld until actual costs and fees are reconciled. Schemas, validators and SDK constructors ship in 0.3.0; assignment, analytics and payment runtimes remain the integrating service's job.