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 skills add budagov-lab/DreamTeam --skill researcher-context-compressiongit clone --depth 1 https://github.com/budagov-lab/DreamTeamWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/budagov-lab/dreamteam/researcher-context-compression)<a href="https://agentmods.dev/skills/budagov-lab/dreamteam/researcher-context-compression"><img src="https://agentmods.dev/badge/skills/budagov-lab/dreamteam/researcher-context-compression/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/budagov-lab/dreamteam/researcher-context-compression"><img src="https://agentmods.dev/badge/skills/budagov-lab/dreamteam/researcher-context-compression.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00041 | $0.00335 |
| Opus 5 | $0.00020 | $0.00168 |
| Sonnet 5 | $0.00008 | $0.00067 |
| Haiku 4.5 | $0.00004 | $0.00034 |
Grade A, and why
researcher-context-compression 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 10d 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.
What it actually says
Researcher Context Compression
When to Use
- When
python -m dreamteam update-task <id> doneprintsTRIGGER_RESEARCHER(source of truth) - After every 20 completed tasks
- When context feels noisy or architecture is unclear
Workflow
- Read from DB: Terminal →
python -m dreamteam memory-get summaries,python -m dreamteam memory-get architecture,python -m dreamteam recent-tasks 20(or MCP dreamteam_*) - Summarize: Condense last 20 tasks into a brief summary
- Update architecture: Add new modules, dependencies, ownership
- Compress: Remove redundant or outdated content (see Researcher agent compression rules)
- Write to DB: Draft to
.dreamteam/temp/, then Terminal →python -m dreamteam memory-set summaries <file>,python -m dreamteam memory-set architecture <file>(or MCP dreamteam_set_memory)
CRITICAL: Read and write memory ONLY via DB (memory-get, memory-set). Do NOT read/write .dreamteam/memory/*.md directly.
Output
- Updated summaries and architecture in DB (via memory-set)
- Orchestrator runs memory-to-files after Researcher
Rules
- Keep summaries concise
- Preserve critical architectural decisions
- Document module → owner mapping
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.
- 10d ago First seen · 34 lines · 41 tokens per session scan A 534afd7fb774
researcher-context-compression is a skill published in the GitHub repository budagov-lab/DreamTeam (1 stars, last pushed 5mo ago), licensed MIT. It adds 41 tokens to every session and 335 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
session-state
Track implementation decisions and progress in specs/state.yaml to prevent context rot. Use at the start of a session to load context, and whenever a significant decision is made or a milestone is reached.
conversation-memory
Persistent memory systems for LLM conversations including short-term, long-term, and entity-based memory.
data-structure-protocol
Give agents persistent structural memory of a codebase — navigate dependencies, track public APIs, and understand why connections exist without re-reading the whole repo.
ejentum-reasoning-harness
MCP server exposing four cognitive harness modes (reasoning, code, anti-deception, memory). Each call returns an engineered scaffold (failure pattern, procedure, suppression vectors, falsification test) the agent ingests before generating.
geminiignore-finops
Configure and optimize .geminiignore files for AI context window efficiency and token cost reduction (FinOps).
hierarchical-agent-memory
Scoped CLAUDE.md memory system that reduces context token spend. Creates directory-level context files, tracks savings via dashboard, and routes agents to the right sub-context.