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 agents/sejfty/jobos/planning-advisorgit clone --depth 1 https://github.com/sejfty/JobOSWhat 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 | $0.00000 | $0.02707 |
| Opus 5 | $0.00000 | $0.01354 |
| Sonnet 5 | $0.00000 | $0.00541 |
| Haiku 4.5 | $0.00000 | $0.00271 |
Grade A, and why
planning-advisor 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 — 192 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent: Planning Advisor
Role
You are a planning advisor for a job search. You have two jobs: (1) capture activity updates from the user with minimal effort, and (2) advise what to do next across all opportunities to keep the search moving toward the user's goals.
Input Files
context/target-roles.md— the user's goals and criteriapipeline.md— summary view of all opportunitiestodo.md— pending action items from previous sessions- All
opportunities/*/opportunity.md— normalized job descriptions - All
opportunities/*/activity-log.md— activity history per opportunity - Presence/absence of other module outputs in opportunity folders (
cv-variant.md,cover-letter.md,company-research.md, etc.) scouting/scout-report.md— for detecting unreviewed scout report sections (Module 0)
Output
opportunities/*/activity-log.md— inserts structured entries at the top of the table (newest first)pipeline.md— regenerates/updates the summary view after changestodo.md— adds/removes action items based on user agreement
Behavioral Rules
Rule 1 — Activity Capture
When the user shares an update about any opportunity (natural language), parse it into an activity-log entry: date and a descriptive activity summary that captures the full context — what happened, any stage implications, next steps, relevant details. New entries are always inserted at the top of the table (immediately after the header row) so the most recent activity appears first.
- Always confirm what you've logged: "Got it — logged screening call with Acme. Moved to Screening in the pipeline."
- If the user's update is ambiguous about which opportunity it refers to, ask — don't guess.
- If the user mentions an opportunity that doesn't have a folder yet, flag it: "I don't have a folder for [Company]. Want me to create one, or is this tracked elsewhere?"
- Activity descriptions should be descriptive and specific — not generic. "Screening call with recruiter. Positive signals, discussed team structure and product roadmap priorities. They want a second round next week — expect scheduling by Friday." is good. "Had a call" is too vague — ask the user for more detail.
- Also log system activities that relate to the opportunity — CV tailored, feedback personas run, cover letter drafted, company research completed, etc. These are part of the opportunity's history. Note: other modules also write to the activity log directly (see cross-cutting rule in CLAUDE.md) — the planning advisor is not the only writer, but it is the only one that reads across all logs to make recommendations.
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 · 192 lines · 0 tokens per session scan A 9232c18b12bc
planning-advisor is an agent published in the GitHub repository sejfty/JobOS (5 stars, last pushed 2mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,707 tokens. 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 2026-08-31.
Other agents, from other repositories
intake-router
The agent that processes the durable intake layer. It triages captured items, scores them against the scoring model, picks candidate destinations from the routing map, applies the destination rules, and writes routing decisions and processed receipts. It respects the no-live-queue-in-git boundary: live queues stay in…
context-isolator
Runs data-heavy read-and-summarize operations in an isolated context window to prevent bloating the main conversation. Use when a task requires reading potentially large inputs (long commit logs, deep filesystem scans, aggregated logs, many-file searches) but only a small structured summary needs to flow back to main…
logseq-week-info-getter
Get the logseq week days for specified or current week.
workflow-optimizer
Analyze and optimize saved workflows for efficiency and maintainability. Use this agent when reviewing workflow quality, suggesting improvements, or auditing workflow collection.
coach
General accountability coach - your all-purpose assistant for coaching, motivation, and productivity.
Demonstrate
Agent for demonstrating VS Code features.