orchestration

Name: orchestration
Author: gadievron

3.4kstars541forks

How to install orchestration?

gadievron/raptor/orchestration

$npx -y skills add gadievron/raptor --skill orchestration

Installs into the current project.

Summary

Orchestrates multi-agent forensic investigations on public GitHub repositories, coordinating parallel evidence collection, hypothesis formation, verification, and report generation.

Files · 1

View on GitHub

SKILL.md

1	# OSS Forensics Orchestration Skill
2
3	You are orchestrating a forensic investigation on a public GitHub repository.
4
5	## Your Role
6
7	You are the ORCHESTRATOR for OSS forensic investigations. You coordinate evidence collection by spawning specialist agents and managing the analysis workflow. You are the ONLY agent that spawns other agents in this system.
8
9	## Invocation
10
11	You receive: `<prompt> [--max-followups N] [--max-retries N]`
12
13	Default: `--max-followups 3 --max-retries 3`
14
15	Parse these flags from the user's request if present.
16
17	---
18
19	## Workflow
20
21	### Phase 0: Initialize Investigation
22
23	CRITICAL: Run the init script using Bash (this is a pre-approved Bash command):
24
25	```bash
26	source .venv/bin/activate && python .claude/skills/oss-forensics/github-evidence-kit/scripts/init_investigation.py
27	```
28
29	The script will:
30	- Check GOOGLE_APPLICATION_CREDENTIALS (stops if missing)
31	- Create `.out/oss-forensics-{timestamp}/` directory
32	- Initialize empty `evidence.json`
33	- Output JSON with workdir path
34
35	Parse the JSON output to extract the working directory path. You will pass this to all agents.
36
37	If prerequisites fail, STOP and inform user.
38
39	---
40
41	### Phase 1: Parse Prompt & Form Research Question
42
43	Extract from user's prompt:
44	- Repository references (e.g., `aws/aws-toolkit-vscode`)
45	- Actor usernames (e.g., `lkmanka58`)
46	- Date ranges (e.g., `July 13, 2025`)
47	- Vendor report URLs (e.g., `https://...`)
48
49	Form a research question specific enough to produce a report with:
50	- Timeline: When did events occur?
51	- Attribution: Who performed what actions?
52	- Intent: What was the goal?
53	- Impact: What was affected?
54
55	If prompt is ambiguous, use AskUserQuestion to clarify:
56	- Missing repo: "Which repository should I investigate?"
57	- Missing timeframe: "What date range should I focus on?"
58	- Vague scope: "Should I focus on PRs, commits, or all activity?"
59
60	---
61
62	### Phase 2: Parallel Evidence Collection
63
64	Spawn investigators IN PARALLEL using a single message with multiple Task calls.
65
66	IMPORTANT: You MUST spawn these in a SINGLE message to run them in parallel:
67
68	```
69	Task: oss-investigator-gh-archive-agent
70	Prompt: "Collect evidence from GH Archive for <research question>.
71	Working directory: <workdir>
72	Targets: repos=<repos>, actors=<actors>, dates=<dates>"
73
74	Task: oss-investigator-github-agent
75	Prompt: "Collect evidence from GitHub API for <research question>.
76	Working directory: <workdir>
77	Targets: repos=<repos>, commits=<commit_shas>, prs=<pr_numbers>"
78
79	Task: oss-investigator-wayback-agent
80	Prompt: "Recover deleted content via Wayback Machine for <research question>.
81	Working directory: <workdir>
82	Targets: repos=<repos>, urls=<github_urls>"
83
84	Task: oss-investigator-local-git-agent
85	Prompt: "Analyze local repository for dangling commits for <research question>.
86	Working directory: <workdir>
87	Targets: repos=<repo_urls>"
88
89	[CONDITIONAL - only if vendor report URL in prompt]
90	Task: oss-investigator-ioc-extractor-agent
91	Prompt: "Extract IOCs from vendor report for <research question>.
92	Working directory: <workdir>
93	Vendor report URL: <url>"
94	```
95
96	Wait for all agents to complete before proceeding.
97
98	---
99
100	### Phase 3: Hypothesis Formation Loop
101
102	```python
103	followup_count = 0
104	while followup_count < max_followups:
105	# Spawn hypothesis former
106	Task: oss-hypothesis-former-agent
107	Prompt: "Form hypothesis for <research question>.
108	Working directory: <workdir>
109	Evidence summary: <summary of collected evidence>
110	[If retry] Previous rebuttal: <rebuttal content>"
111
112	# Check if agent wrote evidence-request-YYY.md
113	if evidence_request_file_exists:
114	# Read the request
115	evidence_request = read_file(f"{workdir}/evidence-request-*.md")
116
117	# Parse which agent and query needed
118	agent_name = extract_agent_from_request(evidence_request)
119	query = extract_query_from_request(evidence_request)
120
121	# Spawn specific investigator
122	Task: {agent_name}
123	Prompt: "{query}
124	Working directory: {workdir}"
125
126	followup_count += 1
127	continue
128
129	else:
130	# hypothesis-YYY.md was written, break
131	break
132
133	if followup_count >= max_followups:
134	# Inform user that we hit the limit
135	print(f"Reached max followups ({max_followups}), proceeding with available evidence")
136	```
137
138	---
139
140	### Phase 4: Evidence Verification
141
142	Spawn verifier:
143
144	```
145	Task: oss-evidence-verifier-agent
146	Prompt: "Verify all evidence against original sources.
147	Working directory: <workdir>"
148	```
149
150	This produces: `evidence-verification-report.md`
151
152	---
153
154	### Phase 5: Hypothesis Validation Loop
155
156	```pyt

People who liked this also viewed

6 picks

Type

paperclip-create-agentCreate new agents in Paperclip with governance-aware hiring. Use when you need to inspect adapter configuration options, compare existing agent configs, draft a new agent prompt/config, and submit a hSkillsJul 2026157k5
subagent-driven-developmentUse when executing implementation plans with independent tasks in the current sessionSkillsJul 2026154k259k
dispatching-parallel-agentsUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesSkillsJul 2026143k259k
browser-act-skill-forgeForges reusable Skill packages (SKILL.md + scripts) from website exploration via browser-act — no re-exploration later. Use when: user wants a reusable Skill for any website, needs to understand a sitSkillsJul 202685k4.6k
wizardGenerate an interactive bash wizard that walks a human through a manual procedure — third-party setup, a one-off migration, an A→B state transition — opening URLs, capturing values, confirming each stSkillsJul 202675k182k
wonda-cliUsing the Wonda CLI to generate images, videos, music, and audio from the terminal — plus LinkedIn, Reddit, and X/Twitter research and automationSkillsJul 202666k137

orchestration

3.4kstars541forks

View on GitHub

How to install orchestration?

gadievron/raptor/orchestration

$npx -y skills add gadievron/raptor --skill orchestration

Installs into the current project.

Summary

Orchestrates multi-agent forensic investigations on public GitHub repositories, coordinating parallel evidence collection, hypothesis formation, verification, and report generation.

Files · 1

View on GitHub

SKILL.md

1	# OSS Forensics Orchestration Skill
2
3	You are orchestrating a forensic investigation on a public GitHub repository.
4
5	## Your Role
6
7	You are the ORCHESTRATOR for OSS forensic investigations. You coordinate evidence collection by spawning specialist agents and managing the analysis workflow. You are the ONLY agent that spawns other agents in this system.
8
9	## Invocation
10
11	You receive: `<prompt> [--max-followups N] [--max-retries N]`
12
13	Default: `--max-followups 3 --max-retries 3`
14
15	Parse these flags from the user's request if present.
16
17	---
18
19	## Workflow
20
21	### Phase 0: Initialize Investigation
22
23	CRITICAL: Run the init script using Bash (this is a pre-approved Bash command):
24
25	```bash
26	source .venv/bin/activate && python .claude/skills/oss-forensics/github-evidence-kit/scripts/init_investigation.py
27	```
28
29	The script will:
30	- Check GOOGLE_APPLICATION_CREDENTIALS (stops if missing)
31	- Create `.out/oss-forensics-{timestamp}/` directory
32	- Initialize empty `evidence.json`
33	- Output JSON with workdir path
34
35	Parse the JSON output to extract the working directory path. You will pass this to all agents.
36
37	If prerequisites fail, STOP and inform user.
38
39	---
40
41	### Phase 1: Parse Prompt & Form Research Question
42
43	Extract from user's prompt:
44	- Repository references (e.g., `aws/aws-toolkit-vscode`)
45	- Actor usernames (e.g., `lkmanka58`)
46	- Date ranges (e.g., `July 13, 2025`)
47	- Vendor report URLs (e.g., `https://...`)
48
49	Form a research question specific enough to produce a report with:
50	- Timeline: When did events occur?
51	- Attribution: Who performed what actions?
52	- Intent: What was the goal?
53	- Impact: What was affected?
54
55	If prompt is ambiguous, use AskUserQuestion to clarify:
56	- Missing repo: "Which repository should I investigate?"
57	- Missing timeframe: "What date range should I focus on?"
58	- Vague scope: "Should I focus on PRs, commits, or all activity?"
59
60	---
61
62	### Phase 2: Parallel Evidence Collection
63
64	Spawn investigators IN PARALLEL using a single message with multiple Task calls.
65
66	IMPORTANT: You MUST spawn these in a SINGLE message to run them in parallel:
67
68	```
69	Task: oss-investigator-gh-archive-agent
70	Prompt: "Collect evidence from GH Archive for <research question>.
71	Working directory: <workdir>
72	Targets: repos=<repos>, actors=<actors>, dates=<dates>"
73
74	Task: oss-investigator-github-agent
75	Prompt: "Collect evidence from GitHub API for <research question>.
76	Working directory: <workdir>
77	Targets: repos=<repos>, commits=<commit_shas>, prs=<pr_numbers>"
78
79	Task: oss-investigator-wayback-agent
80	Prompt: "Recover deleted content via Wayback Machine for <research question>.
81	Working directory: <workdir>
82	Targets: repos=<repos>, urls=<github_urls>"
83
84	Task: oss-investigator-local-git-agent
85	Prompt: "Analyze local repository for dangling commits for <research question>.
86	Working directory: <workdir>
87	Targets: repos=<repo_urls>"
88
89	[CONDITIONAL - only if vendor report URL in prompt]
90	Task: oss-investigator-ioc-extractor-agent
91	Prompt: "Extract IOCs from vendor report for <research question>.
92	Working directory: <workdir>
93	Vendor report URL: <url>"
94	```
95
96	Wait for all agents to complete before proceeding.
97
98	---
99
100	### Phase 3: Hypothesis Formation Loop
101
102	```python
103	followup_count = 0
104	while followup_count < max_followups:
105	# Spawn hypothesis former
106	Task: oss-hypothesis-former-agent
107	Prompt: "Form hypothesis for <research question>.
108	Working directory: <workdir>
109	Evidence summary: <summary of collected evidence>
110	[If retry] Previous rebuttal: <rebuttal content>"
111
112	# Check if agent wrote evidence-request-YYY.md
113	if evidence_request_file_exists:
114	# Read the request
115	evidence_request = read_file(f"{workdir}/evidence-request-*.md")
116
117	# Parse which agent and query needed
118	agent_name = extract_agent_from_request(evidence_request)
119	query = extract_query_from_request(evidence_request)
120
121	# Spawn specific investigator
122	Task: {agent_name}
123	Prompt: "{query}
124	Working directory: {workdir}"
125
126	followup_count += 1
127	continue
128
129	else:
130	# hypothesis-YYY.md was written, break
131	break
132
133	if followup_count >= max_followups:
134	# Inform user that we hit the limit
135	print(f"Reached max followups ({max_followups}), proceeding with available evidence")
136	```
137
138	---
139
140	### Phase 4: Evidence Verification
141
142	Spawn verifier:
143
144	```
145	Task: oss-evidence-verifier-agent
146	Prompt: "Verify all evidence against original sources.
147	Working directory: <workdir>"
148	```
149
150	This produces: `evidence-verification-report.md`
151
152	---
153
154	### Phase 5: Hypothesis Validation Loop
155
156	```pyt

People who liked this also viewed

6 picks

Type