Orca CLI
stablyai/orca
Operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, skill sharing, worktree comments, and Orca's embedded browser…
Read-only catalog that tells a spawned worker agent its environment variables, temporal date-range skills and how to call each integration.
Install Worker Capability Catalog
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install HarnessMD/munder-difflin capabilities --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .claude/skills && cp -r skills-src/resources/skills/capabilities .claude/skills/capabilities && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .claude/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilitiesType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install HarnessMD/munder-difflin capabilities --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .agents/skills && cp -r skills-src/resources/skills/capabilities .agents/skills/capabilities && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .agents/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install HarnessMD/munder-difflin capabilities --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/resources/skills/capabilities .cursor/skills/capabilities && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .cursor/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/HarnessMD/munder-difflin.git --path resources/skills/capabilities--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install HarnessMD/munder-difflin capabilities --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/resources/skills/capabilities .gemini/skills/capabilities && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .gemini/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install HarnessMD/munder-difflin capabilitiesInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .github/skills && cp -r skills-src/resources/skills/capabilities .github/skills/capabilities && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .github/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add HarnessMD/munder-difflin --skill capabilities -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install HarnessMD/munder-difflin capabilities --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/HarnessMD/munder-difflin.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/resources/skills/capabilities .opencode/skills/capabilities && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "capabilities" agent skill from https://github.com/HarnessMD/munder-difflin/tree/main/resources/skills/capabilities into .opencode/skills/capabilities/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "capabilities", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
capabilities- GitHub stars
- 8.5k
- Token cost
- ~1.4k tokens
- SKILL.md length
- 694 words
- Files
- 1
- Skills in repo
- 7
- Repo updated
- First seen
- Licence
- MIT
At a glance
Read-only catalog that tells a spawned worker agent its environment variables, temporal date-range skills and how to call each integration.
- Works in 4 steps: Your environment → Temporal skills — concrete date ranges,… → Integrations — via the loopback broker → …
- Checking which tools and integrations a spawned worker has
- SKILL.md covers 1. Your environment, 2. Temporal skills — concrete…, 3. Integrations — via the… and 4. Boundaries
- Calls node
What it does
The catalog is written for an autonomous worker spawned by a hive to complete one objective and report back. It lists the injected environment variables (agent ID and name, a private workspace directory with identity, memory, inbox and outbox, and the shared hive root) and explains that messages to other agents go out as one JSON file in the outbox, finishing with a done message that carries a substantive result summary.
The temporal skills resolve concrete date ranges relative to now: today, yesterday, this and last week, last 7 or 30 days, this and last month or quarter, and this and last year, plus a general temporal skill for any window such as last 90 days. Each prints civil dates and half-open UTC instants, using a Node resolver script that only reads the clock. The catalog also mentions external integrations reached through a loopback broker.
When your agent uses it
- Checking which tools and integrations a spawned worker has
- Resolving a relative date range such as last month into exact dates
- Learning how to message the coordinating agent through the outbox
Example prompts
- “What temporal skills and integrations do I have available?”
- “Resolve last30Days into civil dates and UTC bounds for a query.”
- “How do I report my result back to the hive when I am done?”
Requirements
- Node.js to run the temporal resolver script
- Pre-approved tools (allowed-tools): Bash
Workflow steps
4 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit 4321e93. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
Bash
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
node
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Worker Capability Catalog loads about 1.4k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 694 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check: notes
The automated check noted patterns worth knowing about, such as sudo or a known installer.
- NotePre-approves every shell command (allowed-tools: Bash)SKILL.md
allowed-tools: Bash
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from HarnessMD/munder-difflin at commit 4321e93, republished under its MIT licence (© HarnessMD). 694 words, ~1,450 tokens.
Download SKILL.mdSave it as
.claude/skills/capabilities/SKILL.md (or your agent's skills folder).- name
- capabilities
- description
- Your capability catalog — read this at boot. Lists the temporal date-range skills and the external integrations (reached via the loopback broker) available to you as a spawned worker, and exactly how to call each. Read-only. Consult it whenever you're unsure what tools/integrations you have or how to invoke them.
- allowed-tools
- Bash
Worker Capability Catalog
You are an autonomous worker spawned by the hive to do one objective and report back. This catalog tells you what you can do and how to call it — your temporal skills and your external integrations — so you don't have to guess. Everything here is read-only to invoke (the integrations themselves may act, but they are mediated and credential-free from your side).
1. Your environment
The harness injects these env vars (use them; don't hard-code paths):
AGENT_ID,AGENT_NAME— your identity in the hive.AGENT_DIR— your private workspace (identity.md,memory.md,inbox/,outbox/, and.claude/skills/). Your bundled skills live under$AGENT_DIR/.claude/skills/.HIVE_ROOT— the shared hive (PROTOCOL.md, the kanban, other agents).
At boot you also have identity.md (who you are) and HIVE_ROOT/PROTOCOL.md
(the full coordination protocol). To message god or another agent, write ONE
message JSON into $AGENT_DIR/outbox/ (schema in PROTOCOL.md). When finished,
send god an "act":"done" outbox message with a substantive result summary.
2. Temporal skills — concrete date ranges, relative to now
When your task is time-scoped, resolve the dates instead of computing them by
hand. Each skill prints inclusive civil dates (YYYY-MM-DD, your local
timezone) and the half-open [startUtc, endExclusiveUtc) instants for
timestamp queries. All are read-only (clock + stdout only; no writes, no network).
Named shortcuts (invoke directly):
| Skill | Range it resolves |
|---|---|
/today, /yesterday | the single civil day |
/thisWeek, /lastWeek | ISO week (Mon-start); this = Mon→today, last = prior full week |
/last7Days, /last30Days | rolling N-day window ending today |
/thisMonth, /lastMonth | this = 1st→today; last = prior full month |
/thisQuarter, /lastQuarter | this = quarter-start→today; last = prior full quarter |
/thisYear, /lastYear | this = Jan 1→today (YTD); last = prior full year |
For any window — including last90Days, last12Months, or arbitrary
lastNdays / lastNweeks / lastNmonths — use /temporal, or call the
resolver directly:
node "$AGENT_DIR/.claude/skills/temporal/when.mjs" last30Days # one window
node "$AGENT_DIR/.claude/skills/temporal/when.mjs" # all windows
node "$AGENT_DIR/.claude/skills/temporal/when.mjs" --json 90d # JSON only
node "$AGENT_DIR/.claude/skills/temporal/when.mjs" --list # keywordsConvention: this* windows are period-start → today (to-date); last*
named periods are the full prior complete period.
3. Integrations — via the loopback broker
External services are reached through the hive's loopback broker: a local,
authenticated endpoint the harness runs on 127.0.0.1. You call the broker; it
holds the credentials and performs the outbound call on your behalf. You never
see or store a secret, and only loopback callers are accepted — so integration
access is brokered, auditable, and credential-free from your side.
How integrations are surfaced to you is environment-dependent — the exact set enabled depends on how you were spawned and the hive's configuration — so discover what's live at run time rather than assuming. The broker pattern is stable even as specific integrations are added; treat the list below as the current surface, not a fixed contract.
Show full SKILL.md (277 more words)Show less
Concrete capabilities you may have right now (check availability before relying on one):
- Reply to the originating Slack thread. If your objective arrived from
Slack, your dispatch includes the exact loopback reply command —
node "<helper>" --channel <C> --thread <T> --text "<mrkdwn>". Use that command verbatim to post your result back to the thread. Send a substantive reply (a short*bold*headline + the actual outcome/links), never a bare "done". - Hive messaging. Coordinate or hand off by writing an outbox message JSON
to god or another agent (
$AGENT_DIR/outbox/). This is your always-available channel. - Semantic memory (MemPalace). If enabled for you:
mempalace search "<query>"to recall shared knowledge,mempalace wake-upfor a digest. - Enterprise Knowledge Graph. If enabled:
node "$KG_CLI" search "<query>"for ranked passages,node "$KG_CLI" list,node "$KG_CLI" get <id>— use it for company-specific facts instead of guessing. - MCP integrations (filesystem, git, and others) come pre-wired into your session settings when enabled; invoke them as normal tools. The set is gated by the hive's consent configuration.
As additional brokered integrations land (calendar, mail, docs, web fetch, …), they follow the same shape: a brokered, credential-free call discoverable at run time. Pair them with the temporal skills above — resolve the date window first, then pass those concrete ISO bounds to the integration query.
4. Boundaries
- Temporal skills are read-only helpers — they never write or reach the network. The broker mediates all external calls; you hold no credentials.
- Do not push or tag to any remote. Commit locally; god is the sole integrator. Pause only for high-severity actions (remote push, paid/infra changes, deleting something you didn't create) — otherwise work autonomously.
- Finish by reporting to god (
"act":"done") with a real, substantive summary.
© HarnessMD, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in resources/skills/capabilities of HarnessMD/munder-difflin.
Open the folder on GitHubat commit 4321e93
Compare with similar skills
Worker Capability Catalog next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Worker Capability Catalog this skillHarnessMD/munder-difflin | 8.5k | — | ~1.4k | Automated safety check: Notes | MIT | |
| Orca CLIstablyai/orca | 87k | 2 repos | ~593 | Automated safety check: Pass | MIT | |
| Paseo Advisor Second Opiniongetpaseo/paseo | 20k | 1 repos | ~756 | Automated safety check: Pass | Custom licence | |
| O2 Review Loopopenobserve/openobserve | 22k | — | ~3.7k | Automated safety check: Pass | AGPL-3.0 | |
| Rowboat Spaces Team Chatrowboatlabs/rowboat | 18k | — | ~846 | Automated safety check: Pass | Apache-2.0 | |
| Paseo Committeegetpaseo/paseo | 20k | 1 repos | ~496 | Automated safety check: Pass | Custom licence |
Similar skills
Paseo Advisor Second Opinion
getpaseo/paseo
Launches one separate agent through Paseo to give a second opinion on the current task, with a self-contained briefing and no permission to edit files.
O2 Review Loop
openobserve/openobserve
Splits a change into planner, coder and independent reviewer roles: you confirm a spec, a subagent implements it, and a separate reviewer checks each round's local WIP commit.
Rowboat Spaces Team Chat
rowboatlabs/rowboat
Sets the rules for an agent working inside Rowboat Spaces: reply in the thread you were asked in, mention people and agents correctly and hand off work without loops.
Paseo Committee
getpaseo/paseo
Forms a two-agent committee with contrasting profiles to analyze a stuck problem in parallel, reconcile their views and return a consensus plan without editing files.
Mission Control Agent API
builderz-labs/mission-control
Teaches an agent to use the Mission Control dashboard API: register, send heartbeats, fetch assigned tasks, report progress and disconnect, with API key auth.
More from HarnessMD/munder-difflin
Temporal Date Window Resolver
HarnessMD/munder-difflin
Resolves a named time window like lastWeek, last30Days or lastQuarter into exact ISO start and end dates relative to the current moment, with no files written and no network calls.
Read-Only Code Audit
HarnessMD/munder-difflin
Scans the working directory for ignored errors, hard-coded secrets, debt comments, dead exports and type gaps, and reports findings by severity without editing files.
Munder Difflin Hive Sync
HarnessMD/munder-difflin
Runs the start-of-task routine for an agent working in a multi-agent hive: reads memory.md, processes new inbox messages and reminds the agent to record durable facts before ending.
Today Date Resolver
HarnessMD/munder-difflin
Resolve "today" to a concrete ISO date range relative to your run time — today. Returns inclusive civil dates plus exact UTC instants so you have temporal…
Yesterday Date Resolver
HarnessMD/munder-difflin
Resolve "yesterday" to a concrete ISO date range relative to your run time — yesterday. Returns inclusive civil dates plus exact UTC instants so you have…
Md Fetch Summarize
HarnessMD/munder-difflin
Fetch a URL and return a concise markdown summary of its content.
Related
Categories
Questions about Worker Capability Catalog
What does Worker Capability Catalog do?
Read-only catalog that tells a spawned worker agent its environment variables, temporal date-range skills and how to call each integration. The catalog is written for an autonomous worker spawned by a hive to complete one objective and report back. It lists the injected environment variables (agent ID and name, a private workspace directory with identity, memory, inbox and outbox, and the shared hive root) and explains that messages to other agents go out as one JSON file in the outbox, finishing with a done message that carries a substantive result summary.
When should I use Worker Capability Catalog?
Worker Capability Catalog fits situations like: checking which tools and integrations a spawned worker has; resolving a relative date range such as last month into exact dates; learning how to message the coordinating agent through the outbox.
How do I install Worker Capability Catalog in Claude Code?
Run `npx skills add HarnessMD/munder-difflin --skill capabilities -a claude-code`. Or copy the skill folder (resources/skills/capabilities in HarnessMD/munder-difflin) into .claude/skills/capabilities in your project. Claude Code loads it when a task matches its description.
How do I install Worker Capability Catalog in Codex?
Run `npx skills add HarnessMD/munder-difflin --skill capabilities -a codex`. Or copy the skill folder (resources/skills/capabilities in HarnessMD/munder-difflin) into .agents/skills/capabilities in your project. Codex loads it when a task matches its description.
Can I use Worker Capability Catalog in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add HarnessMD/munder-difflin --skill capabilities -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/capabilities, .gemini/skills/capabilities, .github/skills/capabilities and .opencode/skills/capabilities in your project.
What does Worker Capability Catalog need to run?
Going by SKILL.md and its folder, Worker Capability Catalog needs the command-line tools its instructions call (node). Our summary lists: Node.js to run the temporal resolver script. Its frontmatter pre-approves these tools: Bash.
Does Worker Capability Catalog access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Worker Capability Catalog safe to install?
Our automated static check of SKILL.md found notes only (pre-approves every shell command (allowed-tools: bash)), nothing it rates as a warning. It is not a guarantee. Review the folder before installing.
What licence does Worker Capability Catalog use?
Worker Capability Catalog is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Worker Capability Catalog use?
About 1.4k tokens (SKILL.md is roughly 5.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Worker Capability Catalog?
Skills that share tags, products or a category with Worker Capability Catalog: Orca CLI (stablyai/orca, 87k stars), Paseo Advisor Second Opinion (getpaseo/paseo, 20k stars), O2 Review Loop (openobserve/openobserve, 22k stars) and Rowboat Spaces Team Chat (rowboatlabs/rowboat, 18k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Worker Capability Catalog?
HarnessMD (a GitHub organization) maintains it in HarnessMD/munder-difflin, which has 8,527 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on October 6, 2026.
Source: HarnessMD/munder-difflin on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.