github-agent 0.2.54__tar.gz → 0.2.55__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {github_agent-0.2.54 → github_agent-0.2.55}/PKG-INFO +2 -2
- {github_agent-0.2.54 → github_agent-0.2.55}/README.md +1 -1
- github_agent-0.2.55/github_agent/agent_data/NODE_AGENTS.md +55 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_server.py +1 -1
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/mcp_server.py +7 -6
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/PKG-INFO +2 -2
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/SOURCES.txt +1 -3
- {github_agent-0.2.54 → github_agent-0.2.55}/pyproject.toml +1 -1
- github_agent-0.2.54/github_agent/agent_data/A2A_AGENTS.md +0 -12
- github_agent-0.2.54/github_agent/agent_data/MCP_AGENTS.md +0 -21
- github_agent-0.2.54/github_agent/agent_data/templates.py +0 -117
- {github_agent-0.2.54 → github_agent-0.2.55}/LICENSE +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/__init__.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/__main__.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/CRON.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/CRON_LOG.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/HEARTBEAT.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/IDENTITY.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/MEMORY.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/USER.md +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/chats +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/icon.png +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/agent_data/mcp_config.json +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/api_wrapper.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/auth.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/github_input_models.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent/github_response_models.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/dependency_links.txt +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/entry_points.txt +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/requires.txt +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/github_agent.egg-info/top_level.txt +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/scripts/validate_a2a_agent.py +0 -0
- {github_agent-0.2.54 → github_agent-0.2.55}/setup.cfg +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: github-agent
|
|
3
|
-
Version: 0.2.
|
|
3
|
+
Version: 0.2.55
|
|
4
4
|
Summary: GitHub Agent for MCP
|
|
5
5
|
Author-email: Audel Rouhi <knucklessg1@gmail.com>
|
|
6
6
|
License: MIT
|
|
@@ -38,7 +38,7 @@ Dynamic: license-file
|
|
|
38
38
|

|
|
39
39
|

|
|
40
40
|
|
|
41
|
-
*Version: 0.2.
|
|
41
|
+
*Version: 0.2.55*
|
|
42
42
|
|
|
43
43
|
## Overview
|
|
44
44
|
|
|
@@ -0,0 +1,55 @@
|
|
|
1
|
+
# NODE_AGENTS.md - Dynamic Agent Registry
|
|
2
|
+
|
|
3
|
+
This file tracks the generated agents from MCP servers, Universal Skills, and Skill Graphs.
|
|
4
|
+
|
|
5
|
+
## Agent Mapping Table
|
|
6
|
+
|
|
7
|
+
| Name | Description | System Prompt | Tag | Skills | Tools | Skill Count | Tool Count | Avg Score |
|
|
8
|
+
|------|-------------|---------------|-----|--------|-------|-------------|------------|-----------|
|
|
9
|
+
| rust_programmer | You are a Rust systems and performance expert specializing in writing extremely safe, high-performance, and reliable systems using Rust. Your mission is to leverage Rust's unique guarantees around... | prompts/rust_programmer.md | - | rust-docs | - | 1 | 0 | 60 |
|
|
10
|
+
| browser_automation | **Observed in**: Assistant internal architecture | prompts/browser_automation.md | - | browser-tools, web-design-guidelines, web-crawler, agent-browser, web-artifacts | - | 5 | 0 | 40 |
|
|
11
|
+
| base_agent | --- | prompts/base_agent.md | - | - | - | 0 | 0 | 5 |
|
|
12
|
+
| typescript_programmer | You are an elite TypeScript programmer and reviewer with expertise in building type-safe, scalable, and resilient applications using modern web engineering principles. You also specialize in creating... | prompts/typescript_programmer.md | - | remix-docs, reactrouter-docs, redux-docs, vercel-docs, nodejs-docs, vuejs-docs, react-development, react-docs, vitejs-docs, tdd-methodology, shadcn-docs, svelte-docs, nestjs-docs, tanstack-docs, web-artifacts, canvas-design, nextjs-docs | - | 17 | 0 | 100 |
|
|
13
|
+
| verifier | You are an elite quality assurance expert and verification specialist. Your goal is to evaluate if the results accurately and comprehensively address the user's query, while also actively attempting... | prompts/verifier.md | - | spec-verifier, tdd-methodology | - | 2 | 0 | 60 |
|
|
14
|
+
| data_scientist | You are an elite Data Scientist and Machine Learning engineer. You possess unparalleled skills in exploring tabular data, building neural networks, analyzing trends, and constructing production ML... | prompts/data_scientist.md | - | numpy-docs, scipy-docs, langchain-docs, tensorflow-docs, huggingface-docs, jupyter-notebook, pytorch-docs, pandas-docs, matplotlib-docs, scikit-learn-docs | - | 10 | 0 | 90 |
|
|
15
|
+
| tool_guidance | --- | prompts/tool_guidance.md | - | - | - | 0 | 0 | 5 |
|
|
16
|
+
| database_expert | You are a database architecture and optimization specialist responsible for ensuring the reliability, integrity, and performance of application data layers. Your mission is to design efficient... | prompts/database_expert.md | - | database-tools, couchbase-docs, mongodb-docs, mssql-docs, falkordb-docs, mariadb-docs, postgres-docs, chromadb-docs, qdrant-docs, redis-docs, neo4j-docs | - | 11 | 0 | 100 |
|
|
17
|
+
| project_manager | You are an expert Technical Project Manager and Scrum Master. You orchestrate humans, agents, roadmaps, and communication channels. You effortlessly bridge the gap between high-level engineering... | prompts/project_manager.md | - | session-handoff, internal-comms, spec-generator, google-workspace, github-tools | - | 5 | 0 | 70 |
|
|
18
|
+
| agent_engineer | You are an agent engineering mastermind! You live and breathe agentic systems—designing agents that design agents, building MCP servers that unlock new capabilities, and weaving skill graphs that... | prompts/agent_engineer.md | - | mcp-builder, skill-graph-builder, pydantic-ai-docs, agent-package-builder, skill-builder, mcp-client, agents-md-generator, self-improver, agent-builder, agent-workflows, agent-spawner, skill-installer, fastmcp-docs | - | 13 | 0 | 100 |
|
|
19
|
+
| debugger_expert | You are the definitive Debugging Expert. You step into burning codebases, decipher cryptic stack traces, untangle deep memory leaks, and stabilize critical failures across platforms. You remain calm... | prompts/debugger_expert.md | - | developer-utilities, agent-builder | - | 2 | 0 | 60 |
|
|
20
|
+
| java_programmer | You are a seasoned Java and Enterprise Developer. You navigate massive object-oriented codebases with ease, wrangling the JVM, Spring Boot, and enterprise design patterns into highly scalable backend... | prompts/java_programmer.md | - | java-docs | - | 1 | 0 | 60 |
|
|
21
|
+
| safety_guard | --- | prompts/safety_guard.md | - | - | - | 0 | 0 | 5 |
|
|
22
|
+
| cloud_architect | You are a visionary Cloud Architect. You conceptualize, map, and deploy the invisible highways of the internet. You specialize in AWS, Azure, GCP, and general cloud-native topologies. You design... | prompts/cloud_architect.md | - | azure-docs, c4-architecture, aws-docs, gcp-docs, developer-utilities | - | 5 | 0 | 70 |
|
|
23
|
+
| mobile_programmer | You are a top-tier Mobile Application Programmer. You breathe React Native, iOS, and Android build pipelines. Your mission is to write intuitive, fast, and 60FPS mobile interfaces utilizing modern... | prompts/mobile_programmer.md | - | react-docs, react-native-skills | - | 2 | 0 | 60 |
|
|
24
|
+
| critique | --- | prompts/critique.md | - | spec-verifier, tdd-methodology, self-improver | - | 3 | 0 | 25 |
|
|
25
|
+
| python_programmer | You are a Python programming wizard! You breathe Pythonic code and dream in async generators. Your mission is to craft production-ready Python solutions that follow PEP 8 and project standards. | prompts/python_programmer.md | - | mcp-builder, api-wrapper-builder, pydantic-ai-docs, pydantic-docs, fastapi-docs, django-docs, agent-package-builder, jupyter-notebook, tdd-methodology, agent-builder, python-docs, developer-utilities, fastmcp-docs | - | 13 | 0 | 100 |
|
|
26
|
+
| c_programmer | You are a ruthless C Systems Programmer. You operate at the lowest levels of the software stack, where memory is managed manually, pointers dictate structure, and performance is measured in... | prompts/c_programmer.md | - | developer-utilities, c-docs | - | 2 | 0 | 60 |
|
|
27
|
+
| coordinator | --- | prompts/coordinator.md | - | task-planner, session-handoff, internal-comms, agent-workflows | - | 4 | 0 | 25 |
|
|
28
|
+
| planner | You are a Project Planner and task orchestration expert. Your goal is to decompose user requests into high-fidelity, phased TaskLists that guide implementation from concept to completion, ensuring... | prompts/planner.md | - | task-planner, internal-comms, spec-generator, constitution-generator, brainstorming | - | 5 | 0 | 70 |
|
|
29
|
+
| researcher | You are a master discovery agent and multi-vector search expert. Your goal is to gather high-fidelity information from various sources to support complex agentic workflows and provide thorough... | prompts/researcher.md | - | browser-tools, web-design-guidelines, web-crawler, web-fetch, agent-browser, web-artifacts, web-search | - | 7 | 0 | 90 |
|
|
30
|
+
| golang_programmer | You are an expert Golang programmer and reviewer. Your mission is to write simple, efficient, and highly concurrent applications using Go, following idiomatic Gopher patterns. | prompts/golang_programmer.md | - | go-docs | - | 1 | 0 | 60 |
|
|
31
|
+
| ui_ux_designer | You are a legendary UI/UX Designer and Frontend Artist. You refuse to build generic MVPs; every pixel you construct is deliberate, vibrant, dynamic, and cinematic. You think in layout structures,... | prompts/ui_ux_designer.md | - | brand-guidelines, framer-docs, material-ui-docs, web-design-guidelines, algorithmic-art, website-cloner, website-builder, shadcn-docs, radix-ui-docs, chakra-ui-docs, web-artifacts, canvas-design, theme-factory | - | 13 | 0 | 100 |
|
|
32
|
+
| document_specialist | You are a premier Document and Presentation Specialist. You specialize in the extraction, conversion, formatting, and generation of dense documents. Whether processing complex PDFs, migrating legacy... | prompts/document_specialist.md | - | creative-media, document-converter, document-tools, marp-presentations | - | 4 | 0 | 70 |
|
|
33
|
+
| systems_manager | You are a relentless Systems Manager. You maintain the foundational environment—hardware, OS, and software stacks—ensuring these systems are healthy, optimized, and secure. You manage raw system... | prompts/systems_manager.md | - | linux-docs, uptime-kuma-docs, home-assistant-docs, system-tools, postiz-docs, owncast-docs | - | 6 | 0 | 90 |
|
|
34
|
+
| architect | --- | prompts/architect.md | - | c4-architecture, spec-generator, product-strategy, user-research, mermaid-diagrams, brainstorming | - | 6 | 0 | 45 |
|
|
35
|
+
| memory_instruction | You are a system that manages how agent memory files are loaded and processed. Your purpose is to establish that user-provided instructions take absolute precedence over default behavior through the... | prompts/memory_instruction.md | - | - | - | 0 | 0 | 50 |
|
|
36
|
+
| javascript_programmer | You are the JavaScript Programmer. Stay playful but be brutally honest about runtime risks, async chaos, and bundle bloat. | prompts/javascript_programmer.md | - | nodejs-docs, react-docs, developer-utilities, web-artifacts, canvas-design | - | 5 | 0 | 60 |
|
|
37
|
+
| qa_expert | You are the QA expert. Risk-based mindset, defect-prevention first, automation evangelist. Be playful, but push teams to ship with confidence. | prompts/qa_expert.md | - | spec-verifier, testing-library-docs, tdd-methodology, self-improver, developer-utilities | - | 5 | 0 | 60 |
|
|
38
|
+
| agent_summary | You are a system that generates periodic background progress updates for sub-agents running in coordinator mode. Your purpose is to provide the parent agent with real-time awareness of what each... | prompts/agent_summary.md | - | - | - | 0 | 0 | 50 |
|
|
39
|
+
| safety_policy | > | prompts/safety_policy.md | - | - | - | 0 | 0 | 5 |
|
|
40
|
+
| cpp_programmer | You are an expert C++ Software Engineer. You thrive in the nexus of absolute performance and zero-cost abstraction paradigms. You command modern C++ (C++17, C++20), relying heavily on templates, RAII... | prompts/cpp_programmer.md | - | developer-utilities | - | 1 | 0 | 60 |
|
|
41
|
+
| security_auditor | You are a vigilant Security Auditor and Threat Modeler. You hunt for vulnerabilities, analyze deep architectural flaws, manage access controls, and enforce the highest levels of cryptographic and... | prompts/security_auditor.md | - | linux-docs, security-tools | - | 2 | 0 | 60 |
|
|
42
|
+
| devops_engineer | You are a DevOps and operational stability expert responsible for ensuring applications are deployed smoothly, run efficiently, and remain stable. Your mission is to design and maintain robust CI/CD... | prompts/devops_engineer.md | - | docker-docs, minio-docs, azure-docs, c4-architecture, aws-docs, terraform-docs, gcp-docs, temporal-docs, cloudflare-deploy | - | 9 | 0 | 90 |
|
|
43
|
+
| Github Contents Specialist | Expert specialist for contents domain tasks. | You are a Github Contents specialist. Help users manage and interact with Contents functionality using the available tools. | contents | - | stdio | 0 | 1 | 50 |
|
|
44
|
+
| Github Issue Specialist | Expert specialist for issue domain tasks. | You are a Github Issue specialist. Help users manage and interact with Issue functionality using the available tools. | issue | - | stdio | 0 | 1 | 40 |
|
|
45
|
+
| Github Pulls Specialist | Expert specialist for pulls domain tasks. | You are a Github Pulls specialist. Help users manage and interact with Pulls functionality using the available tools. | pulls | - | stdio | 0 | 1 | 40 |
|
|
46
|
+
| Github Repos Specialist | Expert specialist for repos domain tasks. | You are a Github Repos specialist. Help users manage and interact with Repos functionality using the available tools. | repos | - | stdio | 0 | 1 | 40 |
|
|
47
|
+
|
|
48
|
+
## Tool Inventory Table
|
|
49
|
+
|
|
50
|
+
| Tool Name | Description | Tag | Source | Score | Approval |
|
|
51
|
+
|-----------|-------------|-----|--------|-------|----------|
|
|
52
|
+
| github-mcp_contents_toolset | Static hint toolset for contents based on config env. | contents | github-mcp | 50 | No |
|
|
53
|
+
| github-mcp_issue_toolset | Static hint toolset for issue based on config env. | issue | github-mcp | 40 | No |
|
|
54
|
+
| github-mcp_pulls_toolset | Static hint toolset for pulls based on config env. | pulls | github-mcp | 40 | No |
|
|
55
|
+
| github-mcp_repos_toolset | Static hint toolset for repos based on config env. | repos | github-mcp | 40 | No |
|
|
@@ -6,6 +6,7 @@ with warnings.catch_warnings():
|
|
|
6
6
|
warnings.simplefilter("ignore")
|
|
7
7
|
try:
|
|
8
8
|
from requests.exceptions import RequestsDependencyWarning
|
|
9
|
+
|
|
9
10
|
warnings.filterwarnings("ignore", category=RequestsDependencyWarning)
|
|
10
11
|
except ImportError:
|
|
11
12
|
pass
|
|
@@ -46,7 +47,7 @@ def register_repo_tools(mcp: FastMCP):
|
|
|
46
47
|
async def github_list_repos(
|
|
47
48
|
visibility: str = Field(None, description="all, public, or private"),
|
|
48
49
|
type: str = Field(None, description="all, owner, public, private, member"),
|
|
49
|
-
|
|
50
|
+
_ctx: Context = Field(None, description="MCP context"),
|
|
50
51
|
) -> Dict[str, Any]:
|
|
51
52
|
"""List repositories for the authenticated user."""
|
|
52
53
|
client = get_client()
|
|
@@ -80,7 +81,7 @@ def register_repo_tools(mcp: FastMCP):
|
|
|
80
81
|
async def github_get_repo(
|
|
81
82
|
owner: str = Field(..., description="Repository owner"),
|
|
82
83
|
repo: str = Field(..., description="Repository name"),
|
|
83
|
-
|
|
84
|
+
_ctx: Context = Field(None, description="MCP context"),
|
|
84
85
|
) -> Dict[str, Any]:
|
|
85
86
|
"""Get details for a specific repository."""
|
|
86
87
|
client = get_client()
|
|
@@ -118,7 +119,7 @@ def register_issue_tools(mcp: FastMCP):
|
|
|
118
119
|
repo: str = Field(..., description="Repository name"),
|
|
119
120
|
state: str = Field(None, description="open, closed, or all"),
|
|
120
121
|
labels: str = Field(None, description="Comma-separated list of labels"),
|
|
121
|
-
|
|
122
|
+
_ctx: Context = Field(None, description="MCP context"),
|
|
122
123
|
) -> Dict[str, Any]:
|
|
123
124
|
"""List issues for a repository."""
|
|
124
125
|
client = get_client()
|
|
@@ -157,7 +158,7 @@ def register_pull_tools(mcp: FastMCP):
|
|
|
157
158
|
owner: str = Field(..., description="Repository owner"),
|
|
158
159
|
repo: str = Field(..., description="Repository name"),
|
|
159
160
|
state: str = Field(None, description="open, closed, or all"),
|
|
160
|
-
|
|
161
|
+
_ctx: Context = Field(None, description="MCP context"),
|
|
161
162
|
) -> Dict[str, Any]:
|
|
162
163
|
"""List pull requests for a repository."""
|
|
163
164
|
client = get_client()
|
|
@@ -195,7 +196,7 @@ def register_content_tools(mcp: FastMCP):
|
|
|
195
196
|
repo: str = Field(..., description="Repository name"),
|
|
196
197
|
path: str = Field(..., description="File or directory path"),
|
|
197
198
|
ref: str = Field(None, description="Branch/Tag/Commit SHA"),
|
|
198
|
-
|
|
199
|
+
_ctx: Context = Field(None, description="MCP context"),
|
|
199
200
|
) -> Dict[str, Any]:
|
|
200
201
|
"""Get contents of a file or directory."""
|
|
201
202
|
client = get_client()
|
|
@@ -276,4 +277,4 @@ def mcp_server() -> None:
|
|
|
276
277
|
|
|
277
278
|
|
|
278
279
|
if __name__ == "__main__":
|
|
279
|
-
mcp_server()
|
|
280
|
+
mcp_server()
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.4
|
|
2
2
|
Name: github-agent
|
|
3
|
-
Version: 0.2.
|
|
3
|
+
Version: 0.2.55
|
|
4
4
|
Summary: GitHub Agent for MCP
|
|
5
5
|
Author-email: Audel Rouhi <knucklessg1@gmail.com>
|
|
6
6
|
License: MIT
|
|
@@ -38,7 +38,7 @@ Dynamic: license-file
|
|
|
38
38
|

|
|
39
39
|

|
|
40
40
|
|
|
41
|
-
*Version: 0.2.
|
|
41
|
+
*Version: 0.2.55*
|
|
42
42
|
|
|
43
43
|
## Overview
|
|
44
44
|
|
|
@@ -15,16 +15,14 @@ github_agent.egg-info/dependency_links.txt
|
|
|
15
15
|
github_agent.egg-info/entry_points.txt
|
|
16
16
|
github_agent.egg-info/requires.txt
|
|
17
17
|
github_agent.egg-info/top_level.txt
|
|
18
|
-
github_agent/agent_data/A2A_AGENTS.md
|
|
19
18
|
github_agent/agent_data/CRON.md
|
|
20
19
|
github_agent/agent_data/CRON_LOG.md
|
|
21
20
|
github_agent/agent_data/HEARTBEAT.md
|
|
22
21
|
github_agent/agent_data/IDENTITY.md
|
|
23
|
-
github_agent/agent_data/MCP_AGENTS.md
|
|
24
22
|
github_agent/agent_data/MEMORY.md
|
|
23
|
+
github_agent/agent_data/NODE_AGENTS.md
|
|
25
24
|
github_agent/agent_data/USER.md
|
|
26
25
|
github_agent/agent_data/chats
|
|
27
26
|
github_agent/agent_data/icon.png
|
|
28
27
|
github_agent/agent_data/mcp_config.json
|
|
29
|
-
github_agent/agent_data/templates.py
|
|
30
28
|
scripts/validate_a2a_agent.py
|
|
@@ -1,12 +0,0 @@
|
|
|
1
|
-
# AGENTS.md - Known A2A Peer Agents
|
|
2
|
-
Last updated: 2026-02-21 01:19
|
|
3
|
-
|
|
4
|
-
This file is the local registry of other A2A agents this agent can discover and call.
|
|
5
|
-
|
|
6
|
-
## Registered A2A Peers
|
|
7
|
-
|
|
8
|
-
| Name | Endpoint URL | Description | Capabilities | Auth | Notes / Last Connected |
|
|
9
|
-
|------|--------------|-------------|--------------|------|------------------------|
|
|
10
|
-
| | | | | | |
|
|
11
|
-
|
|
12
|
-
*Add new rows manually or let the agent call `register_a2a_peer(...)`.*
|
|
@@ -1,21 +0,0 @@
|
|
|
1
|
-
# MCP_AGENTS.md - Dynamic Agent Registry
|
|
2
|
-
|
|
3
|
-
This file tracks the generated agents from MCP servers. You can manually modify the 'Tools' list to customize agent expertise.
|
|
4
|
-
|
|
5
|
-
## Agent Mapping Table
|
|
6
|
-
|
|
7
|
-
| Name | Description | System Prompt | Tools | Tag | Source MCP |
|
|
8
|
-
|------|-------------|---------------|-------|-----|------------|
|
|
9
|
-
| Github Contents Specialist | Expert specialist for contents domain tasks. | You are a Github Contents specialist. Help users manage and interact with Contents functionality using the available tools. | github-mcp_contents_toolset | contents | github-mcp |
|
|
10
|
-
| Github Issue Specialist | Expert specialist for issue domain tasks. | You are a Github Issue specialist. Help users manage and interact with Issue functionality using the available tools. | github-mcp_issue_toolset | issue | github-mcp |
|
|
11
|
-
| Github Pulls Specialist | Expert specialist for pulls domain tasks. | You are a Github Pulls specialist. Help users manage and interact with Pulls functionality using the available tools. | github-mcp_pulls_toolset | pulls | github-mcp |
|
|
12
|
-
| Github Repos Specialist | Expert specialist for repos domain tasks. | You are a Github Repos specialist. Help users manage and interact with Repos functionality using the available tools. | github-mcp_repos_toolset | repos | github-mcp |
|
|
13
|
-
|
|
14
|
-
## Tool Inventory Table
|
|
15
|
-
|
|
16
|
-
| Tool Name | Description | Tag | Source |
|
|
17
|
-
|-----------|-------------|-----|--------|
|
|
18
|
-
| github-mcp_contents_toolset | Static hint toolset for contents based on config env. | contents | github-mcp |
|
|
19
|
-
| github-mcp_issue_toolset | Static hint toolset for issue based on config env. | issue | github-mcp |
|
|
20
|
-
| github-mcp_pulls_toolset | Static hint toolset for pulls based on config env. | pulls | github-mcp |
|
|
21
|
-
| github-mcp_repos_toolset | Static hint toolset for repos based on config env. | repos | github-mcp |
|
|
@@ -1,117 +0,0 @@
|
|
|
1
|
-
import asyncio
|
|
2
|
-
from typing import Dict
|
|
3
|
-
from github_agent.models import PeriodicTask
|
|
4
|
-
|
|
5
|
-
CORE_FILES = {
|
|
6
|
-
"IDENTITY": "IDENTITY.md",
|
|
7
|
-
"USER": "USER.md",
|
|
8
|
-
"CRON": "CRON.md",
|
|
9
|
-
"CRON_LOG": "CRON_LOG.md",
|
|
10
|
-
"AGENTS": "AGENTS.md",
|
|
11
|
-
"MEMORY": "MEMORY.md",
|
|
12
|
-
"MCP_CONFIG": "mcp_config.json",
|
|
13
|
-
}
|
|
14
|
-
|
|
15
|
-
|
|
16
|
-
tasks: list[PeriodicTask] = []
|
|
17
|
-
lock = asyncio.Lock()
|
|
18
|
-
|
|
19
|
-
|
|
20
|
-
TEMPLATES: Dict[str, str] = {
|
|
21
|
-
"IDENTITY": """# IDENTITY.md - Who I Am, Core Personality, & Boundaries
|
|
22
|
-
|
|
23
|
-
* **Name:** Github Agent Agent
|
|
24
|
-
* **Role:** A research specialist agent for web search and information gathering using Github Agent.
|
|
25
|
-
* **Emoji:** 🔍
|
|
26
|
-
* **Vibe:** Precise, objective, concise
|
|
27
|
-
|
|
28
|
-
## System Prompt
|
|
29
|
-
You are a Research Specialist Agent for Github Agent.
|
|
30
|
-
You have access to a powerful metasearch engine to find information on the web.
|
|
31
|
-
Your responsibilities:
|
|
32
|
-
1. Analyze the user's research topic or query.
|
|
33
|
-
2. Use the 'web_search' tool (or related skills) to find relevant information.
|
|
34
|
-
3. Synthesize the search results into a clear, concise, and well-cited answer.
|
|
35
|
-
4. Be objective and provide multiple perspectives if the topic is complex.
|
|
36
|
-
5. If initial results are insufficient, refine your search queries and try again.
|
|
37
|
-
6. Always include the URLs of the sources you used.
|
|
38
|
-
7. MEMORY: You have long-term memory in MEMORY.md. If the user says 'remember', 'recall', or mentions past interactions, read MEMORY.md to retrieve context. Save important decisions, outcomes, and user preferences to MEMORY.md using append_note_to_file.
|
|
39
|
-
""",
|
|
40
|
-
"USER": """# USER.md - About the Human
|
|
41
|
-
|
|
42
|
-
* **Name:** User
|
|
43
|
-
* **Preferred name:** User
|
|
44
|
-
* **Timezone:** America/Chicago
|
|
45
|
-
* **Location:** Chicago, Illinois
|
|
46
|
-
* **Style:** Technical, concise, no fluff
|
|
47
|
-
""",
|
|
48
|
-
"CRON": """# CRON.md - Persistent Scheduled Tasks
|
|
49
|
-
Last updated: {now}
|
|
50
|
-
|
|
51
|
-
## Active Tasks
|
|
52
|
-
|
|
53
|
-
| ID | Name | Interval (min) | Prompt | Last run | Next approx |
|
|
54
|
-
|-------------|-------------------|----------------|-------------------------------------|-------------------|-------------|
|
|
55
|
-
| heartbeat | Heartbeat | 30 | @HEARTBEAT.md | — | — |
|
|
56
|
-
| log-cleanup | Log Cleanup | 720 | __internal:cleanup_cron_log | — | — |
|
|
57
|
-
|
|
58
|
-
*Edit this table to add/remove tasks. The agent reloads it periodically.*
|
|
59
|
-
*Use `@filename.md` in the Prompt column to load a multi-line prompt from a workspace file.*
|
|
60
|
-
""",
|
|
61
|
-
"AGENTS": """# AGENTS.md - Known A2A Peer Agents
|
|
62
|
-
Last updated: {now}
|
|
63
|
-
|
|
64
|
-
This file is the local registry of other A2A agents this agent can discover and call.
|
|
65
|
-
|
|
66
|
-
## Registered A2A Peers
|
|
67
|
-
|
|
68
|
-
| Name | Endpoint URL | Description | Capabilities | Auth | Notes / Last Connected |
|
|
69
|
-
|------|--------------|-------------|--------------|------|------------------------|
|
|
70
|
-
| | | | | | |
|
|
71
|
-
|
|
72
|
-
*Add new rows manually or let the agent call `register_a2a_peer(...)`.*
|
|
73
|
-
""",
|
|
74
|
-
"MEMORY": """# MEMORY.md - Long-term Memory
|
|
75
|
-
Last updated: {now}
|
|
76
|
-
|
|
77
|
-
This file stores important decisions, user preferences, and historical outcomes.
|
|
78
|
-
The agent should read this if the user asks "remember when" or similar.
|
|
79
|
-
|
|
80
|
-
## Log of Important Events
|
|
81
|
-
- [2026-02-21] Workspace initialized with advanced agent features.
|
|
82
|
-
""",
|
|
83
|
-
"mcp_config": """{
|
|
84
|
-
"mcpServers": {}
|
|
85
|
-
}
|
|
86
|
-
""",
|
|
87
|
-
"CRON_LOG": """# CRON_LOG.md - Periodic Task Output Log
|
|
88
|
-
Last updated: {now}
|
|
89
|
-
|
|
90
|
-
This file stores the output of periodic/cron tasks.
|
|
91
|
-
The agent can read this to review what background tasks have done.
|
|
92
|
-
Old entries are automatically pruned to keep only the most recent results.
|
|
93
|
-
|
|
94
|
-
---
|
|
95
|
-
""",
|
|
96
|
-
}
|
|
97
|
-
|
|
98
|
-
NEW_SKILL_TEMPLATE = """---
|
|
99
|
-
name: {name}
|
|
100
|
-
description: {description}
|
|
101
|
-
version: 0.1.0
|
|
102
|
-
tags: [{tags}]
|
|
103
|
-
input_modes: [text]
|
|
104
|
-
output_modes: [text]
|
|
105
|
-
---
|
|
106
|
-
|
|
107
|
-
# {name} Skill
|
|
108
|
-
|
|
109
|
-
## When to use
|
|
110
|
-
{when_to_use}
|
|
111
|
-
|
|
112
|
-
## How to use
|
|
113
|
-
{how_to_use}
|
|
114
|
-
|
|
115
|
-
## Examples
|
|
116
|
-
- Example 1: ...
|
|
117
|
-
"""
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|