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 instructions/getzikra/zikra/agents-mdgit clone --depth 1 https://github.com/getzikra/ZikraWhat 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 | $0.01106 | $0.01106 |
| Opus 5 | $0.00553 | $0.00553 |
| Sonnet 5 | $0.00221 | $0.00221 |
| Haiku 4.5 | $0.00111 | $0.00111 |
Grade A, and why
Zikra AGENTS.md scanned grade A with 1 finding 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 yesterday.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
4. NEVER use n8n MCP tools for Zikra — use curl only 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.
Zikra Agent Instructions
You have access to Zikra persistent memory. Use it every session.
Setup
Your Zikra configuration is in ~/.claude/CLAUDE.md (or equivalent for your agent). Read it at session start for your webhook URL, token, and default project.
Rules
- ALWAYS search Zikra at the start of every session before doing anything else
- Save decisions to Zikra immediately after making them — not at the end
- Log errors to Zikra when you encounter them so they are trackable
- NEVER use n8n MCP tools for Zikra — use curl only
- Always pass
"tags": nullwhen content_md contains curly braces - Always include
User-Agent: curl/7.81.0in all Zikra curl calls
Key commands
search: {"command":"search","query":"<topic>","project":"<p>","limit":5}
Optional: "memory_type":"decision" — filter results to one type only.
Valid memory_type values: conversation, decision, requirement, error, prompt, note
Omit memory_type to return all types. Invalid value returns a structured error.
Note: include_archived is not currently implemented.
save_memory: {"command":"save_memory","project":"<p>","memory_type":"decision","title":"<t>","content_md":"...","tags":null,"created_by":"<hostname>"}
log_error: {"command":"log_error","project":"<p>","message":"<error message>","context_md":"<detail>"}
# Field name: use 'message' (not 'title'). Sending 'title' is silently ignored.
get_prompt: {"command":"get_prompt","prompt_name":"<n>","project":"<p>","runner":"<hostname>"}
log_run: {"command":"log_run","project":"<p>","runner":"<hostname>","status":"success","output_summary":"<2 sentences>","tokens_input":<n>,"tokens_output":<n>}
# prompt_name is optional — include it only when you fetched a named prompt via get_prompt.
# Ad-hoc sessions (no saved prompt fetched): omit prompt_name entirely (stored as NULL, not an error).
# Named prompt run example:
log_run: {"command":"log_run","project":"<p>","runner":"<hostname>","prompt_name":"g:health_check","status":"success","output_summary":"<2 sentences>","tokens_input":<n>,"tokens_output":<n>}
# Never fabricate a prompt_name. If you did not call get_prompt, leave it out.
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.
- yesterday First seen · 70 lines · 1,106 tokens per session scan A 0ca35c89f9b3
Zikra AGENTS.md is an instructions file published in the GitHub repository getzikra/Zikra (9 stars, last pushed 14d ago), licensed MIT. It adds 1,106 tokens to every session, about $0.0055 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
n8n-mcp AGENTS.md
Instructions for czlonkowski/n8n-mcp, covering repository guidelines, project overview, project structure & module organization, architecture & mcp conventions and build, test, and development commands.
n8n-skills CLAUDE.md
Instructions for czlonkowski/n8n-skills, covering claude.md, project overview, repository structure, the 14 skills and 1. n8n expression syntax.
n8n AGENTS.md
AGENTS.md instructions for n8n-io/n8n, covering agents.md, project overview, general guidelines, agent skills and claude code plugin and essential commands.
selfhost-ai CLAUDE.md
Instructions for kossakovsky/selfhost-ai, covering claude.md, project overview, core architecture, key files and installation flow.
n8n CLAUDE.md
Claude Code instructions for n8n-io/n8n, a project described as: Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.
braindb CLAUDE.md
Instructions for dimknaf/braindb, covering braindb — claude instructions, ⚠ tool priority — read this first, it overrides habit, at the start of every session, 1. get always-on rules (behavioral guidelines) and project structure.