Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/fmflurry/settings-opencode/harness-paritynpx skills add fmflurry/settings-opencode --skill harness-paritygit clone --depth 1 https://github.com/fmflurry/settings-opencodeWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00069 | $0.01405 |
| Opus 5 | $0.00034 | $0.00702 |
| Sonnet 5 | $0.00014 | $0.00281 |
| Haiku 4.5 | $0.00007 | $0.00140 |
Grade A, and why
harness-parity scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Harness Parity — Keep Claude & OpenCode in Sync
When you add, edit, rename, or delete any artifact in .claude/ or .opencode/ (skills, agents, commands, rules, instructions), you MUST replicate the equivalent change in the other harness. This keeps the two harness codebases at parity, avoiding orphaned or divergent tools.
A. Replication Checklist
Follow these steps in order for every harness artifact change:
-
Identify artifact type and source harness
- Artifact type: skill / command / agent / rule-instruction
- Source: which harness changed (Claude
.claude/or OpenCode.opencode/)?
-
Locate or create the mirror path in the other harness
- Use the translation table (Section B) to find the equivalent path.
- Create the directory structure if it doesn't exist.
-
Apply the frontmatter/registration translation
- Consult the translation table for frontmatter fields and required registration entries.
- Copy the body verbatim (for skills, commands, rules, instructions).
- For agents: split the single Claude
.mdinto a prompt.txt+ JSONC config entry.
-
Wire the artifact into the registry/load-path
- Claude: Add an
@-import line to.claude/CLAUDE.mdfor rules/instructions. - OpenCode: Add entries to
opencode.jsonc:instructionsarray,commandobject, oragentobject. - If the artifact needs registration, skipping this step leaves it orphaned.
- Claude: Add an
-
Bring across any reference sub-files
- For skills with reference files (
reference.md,templates.md, etc.), copy them too. - File paths are relative; preserve the directory structure.
- For skills with reference files (
-
For deletes/renames: remove mirrors and their registry entries
- Delete or rename the mirror artifact in the other harness.
- Remove any
@-import,instructionsarray entry, orcommand/agentobject entry referencing it.
-
Validate syntax and references
- OpenCode:
opencode.jsoncstill parses (JSONC allows trailing commas). - Claude: All
@-import paths resolve to existing files. - Cross-check wikilinks (
[[skill-name]]) — they must point to existing skills in the same harness.
- OpenCode:
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 70 lines · 69 tokens per session scan A 861b62563e38
harness-parity is a skill published in the GitHub repository fmflurry/settings-opencode (171 stars, last pushed 20d ago), licensed MIT. It adds 69 tokens to every session and 1,405 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
mcporter
List, auth, and call MCP servers/tools from the terminal.
📝 任务完成后归档
重要提醒: 每次完成复杂调试或开发任务后，主动执行此流程！ 将学到的经验归档为 skill，供以后参考。不要等用户提醒。.
oracle
Best practices for using the oracle CLI (prompt + file bundling, engines, sessions, and file attachment patterns).
agent-mode
Unified tool for managing agent LLM modes (add, remove, update, list, switch).
agento11y-prod-setup
Sets up production evaluation and guardrails for a DEPLOYED AI agent in Grafana Agent Observability, grounded in the agent's own code and its real ingested traffic. The judgment layer on top of the agento11y skill: it reads the agent's source (system prompt, tools, entrypoint) AND samples its live traffic via gcx…
mcp-scripting
Write mcpScript JavaScript for discovering, inspecting, and calling MCP tools.