apec-mcp
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@apec-mcpupdate my job-search preferences"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
APEC MCP Server
MCP (Model Context Protocol) server that exposes the APEC candidate-profile API as tools for any MCP-compatible AI assistant.
API Analysis
API Type
Proprietary REST / JSON over HTTPS.
Not publicly documented. Discovered via browser network capture (HAR files).
Base URL: https://www.apec.fr/cms/webservices/
Authentication
Email/password auto-login — no manual cookie handling required.
The server posts credentials to /.apec-login.do at first use and caches the resulting session cookies in ~/.apec_session.json. On 401/403 the session is automatically invalidated and a fresh login is attempted before retrying.
Env var | Purpose |
| APEC account email |
| APEC account password |
| Numeric account ID (from your APEC profile URL) |
| Numeric profile ID (from your APEC profile URL) |
| Override session cache path (default: |
| Manual cookie override — takes precedence over auto-login (fallback if DataDome blocks login) |
Reflection / Introspection
There is no public API schema (no OpenAPI spec, no GraphQL introspection).
The API was reverse-engineered from three browser sessions:
www.apec.fr.har— initial profile creationwww2.apec.fr.har— portfolio/strengths editingwww3.apec.fr.har— full profile editing (business card, photo upload, reference data)
The backend appears to be a Java Spring application (JSESSIONID, Jahia CMS static assets).
Related MCP server: MCP Employee Identity Server
API Endpoints
Profile
Method | Path | Description |
GET |
| Full profile (the canonical object) |
POST |
| Save experiences + education |
POST |
| Save job-search preferences |
POST |
| Save skills + strengths |
POST |
| Save personal / contact info |
POST |
| Validate before saving |
GET |
| UI display flag |
GET |
| Photo upload paths |
Stats
Method | Path | Description |
GET |
| Total profile views |
GET |
| Days since publication |
GET |
| Views trend (float) |
POST |
| Unread recruiter messages |
POST |
| Saved contacts count |
Search / Autocomplete
Method | Path | Description |
GET |
| Job-title search |
GET |
| General search |
Reference Data
Method | Path | Description |
POST |
| Enum lists (869 entries) |
GET |
| Métier taxonomy node |
GET |
| Country list |
Identification & CV
Method | Path | Description |
GET |
| Current user info |
GET |
| Uploaded CV files |
GET |
| Job-recommendation subscription |
Job Offers
Method | Path | Description |
POST |
| AI-matched offers for the candidate |
GET |
| Full offer details by |
POST |
| Similar/related offers |
GET |
| Public company profile |
Applications & Saved Jobs
Method | Path | Description |
POST |
| Submit an APEC-hosted application |
POST |
| List past applications |
POST |
| Check if already applied to an offer |
POST |
| List saved/bookmarked offers |
Data Model (TypeSpec)
import "@typespec/rest";
import "@typespec/http";
using TypeSpec.Rest;
using TypeSpec.Http;
// ── Shared ──────────────────────────────────────────────────────────────
model Audit {
dateCreation: int64; // epoch ms
dateModification: int64 | null;
utilisateurCreation: string;
utilisateurModification: string | null;
}
// ── Reference / enum entries ─────────────────────────────────────────────
model ReferenceItem {
codePresentation: string;
idNomenclature: int64;
codeNomenclature: string;
idOrganisation: int64;
idOrganisationParent: int64 | null;
libelle: string;
niveau: int32;
ordre: int32;
}
// ── Sub-models ───────────────────────────────────────────────────────────
model Experience {
intitulePoste: string;
idNomFonction: int64;
idNomMetier: int64;
dateEntree: string; // ISO 8601
dateSortie: string | null;
nomEntreprise: string;
expNonCadre: boolean;
numeroOrdre: int32;
audit: Audit;
}
model Formation {
intituleFormation: string;
dateEntree: string; // ISO 8601
dateSortie: string | null;
idNomDiscipline: string;
idNomNiveau: string;
organismeFormation: string | null;
numeroOrdre: int32;
audit: Audit;
}
enum CompetenceType {
LANGUE,
SAVOIR_ETRE,
TECHNIQUE,
METIER,
}
model Competence {
libelle: string;
type: CompetenceType;
idNomCompetence: int64;
idNomNiveau: int64 | null;
miseEnAvant: boolean;
idProfilCadre: int64;
audit: Audit;
id: int64;
}
model AdressePostale {
adresseNumeroEtVoie: string;
adresseCodePostal: string;
adresseVille: string;
adresseBatimentImmResidence: string | null;
adresseComplementAdresse: string | null;
idPays: int64;
audit: Audit;
id: int64;
}
model SouhaitSecteur { idNomSecteurActivite: int64; audit: Audit; id: int64; }
model SouhaitFonction { idNomFonction: int64; idNomMetier: int64; audit: Audit; id: int64; }
model SouhaitContrat { idNomTypeContrat: int64; audit: Audit; id: int64; }
model SouhaitEntreprise{ idNomTailleEntreprise: int64; audit: Audit; id: int64; }
model SouhaitTemps { idNomTempsTravail: int64; audit: Audit; id: int64; }
model SouhaitMode { idNomModeTravail: int64; audit: Audit; id: int64; }
model SouhaitLieu { idNomLieu: int64; distance: int32 | null; audit: Audit; id: int64; }
// ── Full profile ─────────────────────────────────────────────────────────
model ProfilCadre {
id: int64;
idProfilCadre: int64;
idCompteCadre: int64;
idNomStatut: int64;
idNomStatutCandidat: int64;
// Personal info
idNomCivilite: int64;
nom: string;
prenom: string;
dateNaissance: int64; // epoch ms
adressePostale: AdressePostale;
adresseEmail: string;
numeroTelephoneMobile: string;
lienLinkedin: string | null;
photo: string | null;
// Professional summary
metierSouhaite: string;
objectifProfessionnel: string | null;
pointsClesProfessionnels: string | null;
idNomAnneesExperience: int64;
idNomDelaiDisponibilite: int64;
remunerationMinimale: float32;
indicateurMasquerSalaire: boolean;
pretPourRecrutement: boolean;
// Preferences
souhaitsSecteurs: SouhaitSecteur[];
souhaitsFonctions: SouhaitFonction[];
souhaitsContrats: SouhaitContrat[];
souhaitsEnts: SouhaitEntreprise[];
souhaitsTemps: SouhaitTemps[];
souhaitsModes: SouhaitMode[];
souhaitsLieux: SouhaitLieu[];
// Experiences & formations
experiencesCles: Experience[];
formationsCles: Formation[];
// Skills
competences: Competence[];
atouts: string[];
portfolios: unknown[];
// CV
idCvFichier: int64 | null;
// Completion indicators
tauxRemplissage: int32;
indicateurCompletCompetence: boolean;
indicateurCompletCompMea: boolean;
indicateurCompletPortfolio: boolean;
// Versioning
numeroVersion: int32;
numeroVersionCadre: int32;
audit: Audit;
auditCadre: Audit;
}
// ── User identity ────────────────────────────────────────────────────────
model ApecUser {
id: string;
numeroCompte: string;
nom: string;
prenom: string;
email: string;
cadre: boolean;
actif: boolean;
token: string;
sessionId: string;
}MCP Tools
Tool | APEC endpoint | Description |
| GET | Authenticated user details |
| GET | Full profile (use as base for updates) |
| GET | Views, duration, trend |
| GET | Uploaded CVs |
| POST | Unread recruiter messages |
| POST | Validate before saving |
| POST | Experiences + education |
| POST | Job-search preferences |
| POST | Skills + strengths |
| POST | Personal / contact info |
| GET | Job-title autocomplete |
| GET | General search |
| GET | Métier taxonomy |
| POST | Enum reference data |
| POST | AI-matched job offers (paginated) |
| GET | Full offer details |
| POST | Related offers |
| GET | Public company profile |
| POST | Submit APEC-hosted application |
| POST | Past applications (paginated) |
| POST | Check if already applied |
| POST | Saved/bookmarked offers |
| POST | Force re-login and return user info |
Update pattern
All update endpoints require the complete profile object (not a partial patch).
profile = get_profile()
profile["experiencesCles"].append({...})
update_experiences_formations(profile)Key Use Case — Fill Profile from LinkedIn or CV
The intended end-to-end workflow:
1. [LinkedIn MCP] get_my_profile() → raw LinkedIn data
OR parse CV text (PDF / plain text)
2. [APEC MCP] get_reference_lists([…]) → resolve enum IDs
search_metiers(q=…) → find idNomMetier / idNomFonction
3. AI assistant maps fields:
LinkedIn title → metierSouhaite + experiencesCles[].intitulePoste
LinkedIn positions → experiencesCles[]
LinkedIn education → formationsCles[]
LinkedIn skills → competences[]
LinkedIn location → adressePostale + souhaitsLieux[]
4. [APEC MCP] get_profile() → fetch current profile
update_experiences_formations(…) → save
update_informations_complementaires(…)
update_atouts(…)
update_carte_de_visite(…)Setup & Run
Prerequisites
uv installed
Install
git clone <repo>
cd apec-mcp
uv sync
cp .env.example .env
# Fill in APEC_EMAIL, APEC_PASSWORD, APEC_ID_COMPTE_CADRE, APEC_ID_PROFIL_CADRELocal development (SSE)
uv run apec-mcp # SSE on http://localhost:8080/sseMCP client config (stdio — recommended)
Pass credentials directly in the env block. Use the absolute path to the venv binary so the config works regardless of which directory the MCP client opens.
{
"mcpServers": {
"apec": {
"command": "/path/to/apec-mcp/.venv/bin/apec-mcp",
"args": [],
"env": {
"MCP_TRANSPORT": "stdio",
"APEC_EMAIL": "you@example.com",
"APEC_PASSWORD": "yourpassword",
"APEC_ID_COMPTE_CADRE": "123456789",
"APEC_ID_PROFIL_CADRE": "1234567"
}
}
}
}For Claude Code place this block in ~/.claude.json under mcpServers so all sessions share it.
Docker (SSE mode, optional)
docker build -t apec-mcp:latest .
docker run -d --name apec-mcp-server -p 8081:8080 --env-file .env \
-e MCP_TRANSPORT=sse apec-mcp:latestThen point your MCP client at http://localhost:8081/sse.
docker composemay fail to resolve build-cache image digests in some container-in-container environments (OrbStack socktainer, Docker-in-Docker). Plaindocker runis more reliable in those setups.
Session Refresh
Session cookies are cached automatically in ~/.apec_session.json and refreshed on 401/403. No manual action is normally needed.
If auto-login is blocked by DataDome:
Open
https://www.apec.fr→ log in manuallyDevTools → Network → any
/cms/webservices/*request → Request Headers →CookieAdd
APEC_COOKIES=<value>to.env— this overrides auto-login
This server cannot be deployed
Maintenance
Related MCP Connectors
- UnifAPIOAuthcom.unifapi
Hosted MCP server for live public-data APIs and Skills for AI agents.
MCP server that lets AI assistants use all OneSchema features exposed via the public API.
MCP server exposing the Backtest360 engine API as tools for AI agents.
GetJobzi MCP server for job search, application tracking, and career forecasting.
Related MCP Servers
AlicenseNot gradedqualityAmaintenanceMCP server for job search and application tracking, enabling AI agents to search jobs, get details, manage applications, and find contacts across 128K+ jobs and 1,900+ companies.638 npm3MIT- AlicenseNot gradedqualityDmaintenanceThis MCP server enables AI-assisted professional profile generation by connecting to verified employee work data (skills, career history, certifications, projects) from employer HCM systems.MIT
- AlicenseNot gradedqualityCmaintenanceAn MCP server that exposes job-search and application-management capabilities to compatible AI clients, enabling discovery of vacancies, drafting of tailored application materials, and coordinated human-approved submissions.MIT
- FlicenseAqualityBmaintenanceMCP server that exposes a resume as callable tools and resources, enabling AI agents to query experience, skills, projects, and contact information via natural language.3-